JRDF is very crap (part 2)

It didn't take as long as expected which means it's probably wrong. For non-filtered queries it's three times faster and for certain FILTER queries (with equals) it's 47 times faster (from 284 to 6 seconds). At least it's now in the same order of magnitude as most tools and it's a a tiny bit faster than some (although adding more features will probably slow it down again).

What changed:
* AttributeValuePair has been removed and replaced with maps (as discussed previously).
* Seeing as though maps were used so much hashCode and equals were optimized. As I've found before (I think), isAssignableFrom is slower than try/catch for equals (depending on your usage of course).
* Queries go through unsorted and uncopied rather than standard graph finds. I'd forgotten about how much effort had gone into allowing remove and automatic sorting on iterators.
* A very simple optimizer (it's really only simplifying the FILTER constraints at the moment) was added. Tree manipulation was painful - I resorted to mutating in place operations.
* Better designed. It's a bit hard to qualify this except what was there was truly awful - objects being created in constructors and passing itself in. The nice thing about IoC is it's quite easy to see when you're not using objects at the same architectural level.

Update: For download.

JRDF is very crap (part 1)

I've been spending some time looking at the querying part of JRDF. And it's quite bad. How bad? Well I've been profiling it and noticing that a lot of time was spent comparing attribute value pairs. An attribute in JRDF consists of a name (variable or position in a triple) and type (position in a triple or literal, URI Reference or blank node). Comparisons are done during most operations (like joins) and they are done on sorted attribute values. This is incredibly dumb. What's much better is to have a map of attributes to values. No sorting required and O(1) lookup - hurrah. The code around the comparisons also got a lot simpler and is obviously better. I think there's at least one other case of this at a different level and potentially room for about an order of magntitude speed up over the current release. Test queries are already 2-3 times faster.

The main reason for this though, is that currently the FILTER in JRDF runs about 10 times more slowly than a query using triple matching (this isn't a complexity measurement - it's based on a rather small set of triples). So a query with "?a <some:value> 'foo'" is much slower than "?a <some:value> ?b . FILTER(str(?b) = 'foo')". The queries aren't the same but their performance shouldn't be that much slower. However, in order to get to a stage of improving FILTER's performance the code has to be refactored - hopefully simpler and faster.

FILTER is nicely functionaly - it seems a shame to implement it in Java - it's eye poppingly bad at the moment. I was thinking functional Java but instead of taking the gateway drug I was think of just going to the hard stuff. FILTER is operated by creating different operations within a relation - which allows you to put ANDed FILTERs vertically across a relation (columns) and ORed FILTERs horizontally (rows). I don't know if anyone else implements it this way - it might be another bad idea over time.

Mice Spiders

Interview with Simon Pegg on Spaced. The UK DVDs are still the only source of the show (and ABC2) for Australians until October.

YADS and RDF Molecules

BNodes Out! discusses how any usefully scalable system doesn't use blank nodes. What is interesting is the comment on YADS (Yet Another DOI Service). The best reference is Tony's presentation although it is mentioned in Jane's as well. "YADS implements a simple, safe and predictable recursive data model for describing resource collections. The aim is to assist in programming complex resource descriptions across multiple applications and to foster interoperability between them...So, the YADS model makes extensive use of bNodes to manage hierarchies of “fat” resources - i.e. resource islands, a resource decorated with properties. The bNodes are only used as a mechanism for managing containment."

This sounds a lot like RDF molecules and supports visualization (apparently). This seems like a good use of molecules that I hadn't previously thought of (Tony's talk gives an example of the London underground). The main homepage of YADS isn't around anymore - it'll be interesting to see if it's still being used/worked on.

Update: Tony has fixed up the YADS home page (there's also an older version).

Hadoop and Microsoft

Pluggable Hadoop lists some extensions to Hadoop in the pipeline: job scheduling (including one based on Linux's completely fair scheduler), block placement, instrumentation, serialization, component lifecycle, and code cleanup (the analysis used Structure101).

I found the reason why HQL was removed from HBase (to be replaced by a Ruby DSL and to ensure that HBase wasn't confused with an SQL database) and moved to HRdfStore.

There's also rumours that Microsoft's recent investment in Apache may lead to them working on Hadoop too.

Save us China

I was in Victoria when the ETS for Australia was announced (well the discussion papers). It's fairly funny, that replacing the world's worst plants even with other coal plants (Hazelwood is the world's worst), with Chineese brown coal plant technology, would reduce emissions by 30% to 40% (by just drying out the brown coal). It's still very poluting but it just shows how far behind Australia is. This has lead to greater compensation to Victorian polluters (which is just mad). At the same time Queensland is creating another coal port because we can't export the carbon fast enough.

The exclusions were annoying (aluminium, cement and some types of steel). Cement is annoying (5% of all CO2 apparently) as there exists green alternative technologies. The time is to invest not compensate.

Square brackets are scary

For what may be an increasing trend of surfing the Web at 320x480 I noticed Cydia has a number of applications for Jailbroken iPhones (Java, Python and Ruby mainly). The mailing list on iPhone/Java doesn't have much on it except some interesting uses of JocStrap and UICaboodle (available from SVN by Jay Freeman). There's also the Sun blog that has some interesting sample applications using different Java implementations on the iPhone.

JRDF 0.5.5.1 Released

Just a quick note about a new version of JRDF. It's been a short time between releases but it still contains one significant advance over the previous one and that's persistent graphs. It's still in the early stages but it's basic enough for simple use cases. It also contains text serialization (based on NTriples) that is useful for moving RDF molcules around nodes in a cluster (for example). A lot of this code is fairly much "spike" code and I expect that another release will be released after we exercise these new features more (and write some tests/rewrite the code).

Update: 0.5.5.2 is now available fixing many bugs and introducing FILTER support.

Of Mats and Cats

No universal things Re: comparing XML and RDF data models was started by Bernard Vatant. This comes to the heart of whether people can know reality (well that's how I'd summarize the idea of universals see Beyond Concepts).

There were a few quotes that I found interesting:

It's been counter-productive in science for centuries. Physics had to go over the notion of universal thing to understand that light is neither a wave, nor a particle. Biology to go over the notion of taxa as rigid concepts based on phenotypes to understand genetics etc. Many examples can be found in all science domains. My day-to-day experience in ontology building, listening to domain experts, is indeed not that 'there are things that people are trying to describe', but that 'there are descriptions people take for granted they represent things before you ask, but really don't know exactly what those things are when you make them look closely'.

Bijan wrote:

I do think that the family of views in computational ontologies generally called "realist" is indeed naive and fundamentally wrong headed. Whether it's a "useful fiction" that helps people write better or more compatible ontologies is an open empirical question.

But I, for one, wouldn't bet on it.

I remember also a project where we were trying to get people to write simple triples. They got that they needed triples. But what they ended up putting into the tool was things like

S P O
"The cat is" "on the" "mat".
"Mary eats" "pudding" "on toast"

They just split up the sentences into somewhat equal parts!

I really feel like an interested amateur and my view is probably influenced by databases in computer science, where you are taking the non-realist approach. I say this because there are usually properties in databases that are not really based on reality but are a result of other requirements (like a column like "isDeleted" rather than actually deleting the statement).

Round of Links

• Apache Hadoop Wins Terabyte Sort Benchmark "One of Yahoo's Hadoop clusters sorted 1 terabyte of data in 209 seconds...This is the first time that either a Java or an open source program has won." There were just under 1000 nodes, the benchmark results are hosted by HP (a tad more detail here).


• Microsoft buys Powerset one of the interesting things is that they use Hadoop (see their blog). It's hard to tell whether this is bad or good for Hadoop.


• Google vs Microsoft - oh for structure.


• Tom talking about GridGain from his presentation in February. C++ isn't as productive as Java?


• Applets are back (according to Sun).


• Why commenting is for n00bs. "And Haskell, OCaml and their ilk are part of a 45-year-old static-typing movement within academia to try to force people to model everything. Programmers hate that. These languages will never, ever enjoy any substantial commercial success, for the exact same reason the Semantic Web is a failure. You can't force people to provide metadata for everything they do. They'll hate you."


• Some interesting discussion on Web 2.0 and the future of the web.


• Rich text editor for browsers. Not free though.


• Linked data and what it is.


• ThoughtWorks Podcasts (the REST talk was what drew me to it).


• Turtle specification. I've been looking at this for serialization of RDF molecules but it seems that you can't have blank nodes as objects using the nested syntax.


• Semantic Web for bioinformatics.


• Data structure stuff: Linear Bloom Filters, Bloom filters for Spell Checking, Optimal Bloom Filter replacements and scalable btree and B-tries for Disk-based String Management.

Ob. iPhone 2

Good to see carriers actually putting up a bit of a fight for iPhone business. Telstra announces iPhone 3G details with $279, $30 a month on a 24 month, with free access to WiFi hotspots. This better be true.

Update: Optus releases pricing

ScalaCC

Formal Language Processing in Scala which links to External DSLs made easy with Scala Parser Combinators that I'd read from here.

Although, just to keep it balanced I have noticed Steve Yegge's comments, under "Static Typing's Paper Tigers", on the complexity of Scala's typing (it does have a lot) and it has been pointed out that this does lead to problems with writing IDEs to support it.

JRDF 0.5.5

The main difference in version 0.5.5 from the previous one is the inclusion of a RDF molecule store. Both in memory and disk based versions are supported and can be queried just like a normal triple store. This is also the first version that has been renamed URQL instead of SPARQL for the query evaluation. The SPARQL grammar is the same but it does not support the weird outliers that SPARQL has for empty graph patterns but follows relational (and other) algebras. There's also the usual bug fixes and other features.

Update: Due to a couple of bugs found in 0.5.5 there will be a 0.5.5.1 version released soon.

Beef of the Sea

Everyone is probably sick of me talking about the Gruen Transfer. So what better what to continue to talk about it than to blog about it. Perhaps the best part of the show is The Pitch especially episode one's selling whale meat (this is the runner up) and making the Democrats electable (the second is best). Who would've thought deconstructing chocolate adverts would be interesting? One of the good things is that the show is available for download. There is also some good discussion in the forum and links to some other good adverts (although it possible should've been crows).

Bad Balmer

Eight Years of Wrongness. Lists some of the things believed to have gone wrong with Microsoft in the last 10 years or so. They include: losing the DOJ and EU cases, Vista, XBox, IE, Zune, and Windows Mobile. Linked mainly because they use Fake Steve as a source of analysis.

Apple Sprouts

AppleInsider has some details on SproutCore. The official web site says, "makes Javascript fun and easy" - and it's just a Ruby gem install away. They also link to some previous talk about Cocoa for Windows.

Apple's trojan horse in the runtime wars has been well known for a while.

The photo demo looks a lot like the MobileMe Gallery that was presented at WWDC 2008 (SproutCore doesn't seem to work too well under IE 7 and the rotation only works in Safari). Gallery has less functionality than things like Photoshop Express although the integration is obviously better.

There's also an interesting Javascript library for drawing 2D objects (UML, workflows, etc) that I've been shown recently called Draw 2D.

The Curse of the Floppy Penises

A Western floppy penis is more valuable than preventing blindness in an African eye (see neglected diseases). This is part of the story in the video of the launch of "The Health Commons". The video talks about how hundreds of thousands of people go blind from "river blindness". It has very little value associated with it and drug companies focus on more valuable drugs to do with baldness and erectile disfunction. The video goes on to talk about how the network changes things and how there's a lack of process change in science to take advantage of these effects. If you can leverage network effects then this hopefully reduces the cost of drug discovery making drug development in less valuable diseases viable. The white paper covers some more of this in detail.

It also talks about an idea that I've often thought of as useful - the collection of failed experiments, "This deeply set inability to capture collective learning dooms everyone to revisit infinitely many blind alleys. The currency of scientific publication encourages individual scientists to hoard rather than share data that they will never have the time or resources to exhaustively mine. And, the wealth of “negative” information gleaned from clinical trial data is mostly lost to the need for companies to safeguard their commercial investments."

The general idea seems to share and standardize all aspects of research and science.

Ob. iPhone

So I've been trying to find more information from a variety of sources on pricing.

The closest to reality that I've been able to find is these leaked details from Optus (via Gizmodo):
"The iPhone will only be available on a 24 month contract – no outright purchase, with the 8GB model to sell at AUD $220, and the 16GB model at $330, with only the 16GB model in white as Steve Jobs announced at the WWDC keynote.

Accessories will only be available through Apple stores – Optus will only carry the iPhone 3G itself, and the all important voice and data plans are as follows: $79 cap for $300 worth of calls and 1GB of data, or a $99 cap with $400 worth of calls and a 3G data download limit.

Visual voicemail is included, and the cap is whittled away in 35c per 30 second chunks, 25c per SMS message and the always annoying but always present flagfall which is set at 30c."

This makes it over twice as expensive as the ATT plans (and I think they had unlimited data). This is where I get cranky about Australian carriers and their stupid plans. It would probably count me out at those prices.

Update: No more Apple rumours. As Brad says in the comments, this is wrong.
Update 2: Looks like the UK is getting a good deal.
Update 3: Gizmodo link gone...nothing to see here.

Evolution and SUVs

Two things that have I've been interested in before: SUVs and evolution.

The first mirrors what is happening in Australia too where small cars are winning over larger ones and Falcon sales have dropped by half. The F150 isn't really that popular here.

The observation that bacteria have evolved to process other nutrients is interesting. But the related articles tend to be a bit more forceful: such as the Bible's many inaccuracies, the many occurrences of homosexuality and how it actually works in nature and that accepting evolution does not mean rejecting morality.

Linked Data, FOAF, and OWL DL

So I spent a little time a while ago looking through all the different ways ontologies support linked data. Some of my data I wish to link together is not RDF but documents that define a subject. For example, a protein will have peer reviewed documents that define it. It's not RDF but it is important.

The tutorial on linked data has a little bit of information: "In order to make it easier for Linked Data clients to understand the relation between http://dbpedia.org/resource/Alec_Empire, http://dbpedia.org/data/Alec_Empire, and http://dbpedia.org/page/Alec_Empire, the URIs can be interlinked using the rdfs:isDefinedBy and the foaf:page property as recommended in the Cool URI paper."

The Cool URIs paper, Section 4.5 says: "The rdfs:isDefinedBy statement links the person to the document containing its RDF description and allows RDF browsers to distinguish this main resource from other auxiliary resources that just happen to be mentioned in the document. We use rdfs:isDefinedBy instead of its weaker superproperty rdfs:seeAlso because the content at /data/alice is authoritative."

There is also some discussion about linking in URI-based Naming Systems for Science.

Now my use case is linking things to documents that define that thing. So rdfs:seeAlso is not appropriate as it "might provide additional information about the subject resource". And rdfs:isDefinedBy is also out as it is used to link RDF documents together. I need a property that defines a thing, is authoritative but isn't linking RDF (it's for humans). I also would like to keep my ontology within OWL DL.

FOAF has a page property. I've used the OWL DL version of FOAF before and FOAF cleaner (or should that be RDFS cleaner). So it seemed like a good match. However, its inverse is topic which isn't good. Because I'm linking the thing to the page - it's not a topic. So scrub that.

RSS has a link property which extends Dublin Core's identifier. This seems more like it. However, I'd like to extend my own version of link and I'm stuck because as soon as you use RDFS vocabularies in OWL DL you're in OWL Full territory. It'd be nice to stay in OWL DL. There is an OWL DL version of Dublin Core. All of the Dublin Core properties are nicely converted to annotation properties. However, you're still stuck because you can't make sub-properties without going into OWL Full. I like the idea of annotation and semantically Dublic Core seems to be a suitable vocabulary of annotation properties. Extending Dublin Core is out of OWL DL - which is shame because it's probably the closest match to what I wanted.

As an aside, annotation properties are outside the reasoning engine. The idea is that you don't want an OWL reasoner or RDF application necessarily inferring over this data or trying to look it up in order for the document to be understood. So the way they do it in OWL DL is to have annotation properties that are outside of/special to the usual statements. Sub-properties require reasoning, so limiting them makes some sense but it does hamper extensibility - it'd be nice to express them and turn on the reasoning only when asking about those properties (I think Pellet has this feature but I didn't look up the details).

The other vocabulary I looked at was SIOC's link. Again, this seems like a close match but again it's RDFS.

In the end, I just created another annotation property called link.

In summary:

• For my requirements, the suggestions for linking data seems to only work for RDF and RDFS ontologies. Reusing RDFS from OWL DL or OWL DL from RDFS doesn't look feasible as one isn't a subset of the other (an old problem I guess).


• Current, popular Semantic Web vocabularies are in RDFS. Why aren't there more popular OWL DL versions of these things? Is the lack of extensibility holding it back?


• Is my expectation wrong - should I stick within OWL DL or is an RDFS and OWL DL combination okay?


• Why not allow annotation properties to have sub-properties?


• Maybe the OWL DL specification does have suitable properties for linking certain data but I don't understand which is the right one.

Update: The Neurocommon's URI documentation protocol is quite similar as well. Except that, it seems to be too specific as it ties the name with a single thing that defines it. All the parts of Step 5 could potentially be eliminated with what I'm thinking of.

Somewhere

Alarm Bells Sound for the Amazon

Brazil's land mass and farming industry make it one of the most agriculturally productive countries in the world. It has already been dubbed "the world's feeding bowl" and is exporting more and more to emerging economies, such as India and China.

As China's middle-class continues to grow, so, too, does its demand for food. Brazil exports 10 million tons of soybeans to China a year for both animal feed and human consumption, trade that is crucial to Brazil's economic development.

And it's not just poverty that's an issue.

The state of Para has some of the worst human rights abuses in Brazil. People are trafficked from across the impoverished northeast of the country to work in slavelike conditions in the sawmills, illegal charcoal ovens and cattle farms.

They usually work in horrific conditions, with no basic rights and existing on roughly $5 a day. If they try to seek help from the authorities, they are threatened with death.

There's also the WHO page on "Deaths from Climate Change".

RDF Processing

One of the interesting things about biological data, and probably other types, is that a lot of it is not quite the right structure. That's not to say that there's not people working to improve it, the Gene Ontology seems to be updated almost daily, but data in any structure maybe wrong for a particular purpose.

Biologists make a habit, out of necessity, of just hacking and transforming large amounts of data to suite their particular need. Sometimes, these hacks get more generalized and abstracted like GO Slims. We've been using GO Slims in BioMANTA for sub-cellular location (going from 2000 terms to 500). GO contains lots and lots of information and you don't need it all at once and more often you don't need it at the maximum level of granularity that it has. Some categories only have one or two known instances, for example. You may even need to whittle this down further (from say 500 to 200). For example, when we are determining the quality of an interaction we only care where the proteins exist generally in an organism. If the two proteins are recorded to interact but one is in the heart and the other in the liver then it's unlikely that they will react in the host organism. The part of the liver or the heart and other finer structural detail is not required for this kind of work (AFAIK anyway).

The point is, a lot of our work is processing not querying RDF. What's the difference between the two and what effect does it have?

For a start, querying assumes, at least to some degree, that the data is selective - that the results you're getting is vastly smaller than your original data. In processing, you're taking all of the data or large chunks of it (by sets of predicates, for example) and changing or producing more data based on the original set.

Also, writing is at least as important as reading the data. So data structures optimized for lots of writes, temporary, concurrent, is of greater importance than those built around more familiar requirements for a database.

Sorting and processing distinct items is a lot more important too. When processing millions of data entries it can be quite inefficient if the data has a large number of duplicates and needs to be sorted. Processing can also be decentralized - or perhaps maybe more decentralized.

To top it off, the data still has to be queried. So this doesn't remove the need for efficient, read only data structures to perform selective queries for the usual analysis, reporting, etc. So none of the existing problems goes away.

git + RDF = versioned RDF

Reading, Git for Computer Scientists, and it seems like if you turn the blob into a set of triples you pretty much have versioned RDF (or molecules even).

I'm also wondering, if Digg is so pro-Semantic Web, where's the http://digg.com/semweb?

I See Triples

Digg makes official its adoption of a 'semantic Web' standard "Other brief mentions on Digg's blogs over the past month have been the only indications the company has been giving to the world of its direct -- and perhaps even principal -- involvement in RDF and RDFa, besides a simple check of the site's own source code, where attributions such as rel="dc:source" property="dc:title" within <DIV> elements are now common. A few weeks ago, developer Bob DuCharme discovered these little attributions and began playing with them to discern their viability."

"The possibility exists for a kind of mega-meta-source to emerge from Digg, where interesting news topics are associated with cataloged resources. But for that to actually work, someone has to manage those resources -- and that effort will take a level of humanpower and resources of another kind (the kind symbolized with "$") that RDF won't provide even the most ambitious sites just on its own."

See Digging RDFa. More news about RDFa is available at RDFa.info.

To see Digg in all its RDFa glory one way is to copy this Javascript for highlighting or this one for getting RDF triples into you bookmark bar after the Digg front page has loaded.

So I still haven't finished writing up everything that I've saw at WWW2008 but the overall messages were:

• RDFa is easy and gets people going with RDF quickly (see "They knew the train would come"). Semantic wikis (links to the Semantic Mediawiki project) have also come a long way to making it more err user friendly.


• HTML5 and the end of the browser development winter seems like the death to plugins at last. I hadn't realized this before, but the message seems to be that a plugin is a way of saying to the Web "your browser isn't full featured enough".


• The Facebooks of the world and all those online communities really are a danger to the Web - the creation of data silos. And I'd really like to have the time to write some SIOC plugins to help open up these silos (or just change my blog template to have RDFa).

Bankrupt

"And now, we're [Americans are] the most religious nation on earth - that's why we kill so easily. We're sending people to heaven. And because we are now terribly, terribly religious in a sense that no proper American ever was when I was young - I was in the Second World War." - Gore Vidal.

And they are bankrupt in the finacial sense as well due to Iraq (and other causes of course). The speaker also follows a line I've seen often where the war has been fought without enough commitment from the government (i.e. decreasing taxes instead of increasing them, hiring fighters instead of drafting, etc.). One rather shocking statistic was that 48 percent of returning troops will be disabled in some way - maybe that's because more a living than dying but it's still quite an amazing number - but it means "...we've created just for the disabled in this war in the last five years, a gap equal to the gap that we created over decades in the social security system...It's an order of magnitude worse than the Vietnam War."

When URIs are too Much

Every Subject is a Blank Node "In RDF, URIs are good at defining unambiguous property values, in other words objects, including type. But very often, and maybe most of the time, the individual subject (in both meaning of subject of an RDF triple, and topic maps subject of conversation) is best represented as a blank node bearing all kinds of identified properties, but none of them conferring absolute identity. This way, it's left to applications to figure out identification rules, in other words which property or boolean combination of properties they want to consider as identifying or not."

From the mailing list: "With no URI, you are free to let applications decide which contexts are considered the same or not, based on specific rules on properties. Some applications would decide that all contexts where role "I" is played by "John Black" are the same, and will cluster all contextResource properties, some other will not."

Long tail of programming languages

While I'm sick of long tail blahs, I recently came across the idea that programming languages follow the same power laws found in other areas. This particular long tail this should be encouraging for those who have a disdain for the current mainstream computer languages, "Rather than finding ways to create an even lower lowest common denominator, the Long Tail is about finding economically efficient ways to capitalize on the infinite diversity of taste and demand that has heretofore been overshadowed by mass markets."

Furthermore, "There is a long tail because the more specialized a language is to a domain, the better it fits to solve problems for that domain. These niche languages trade off generality for efficiency in a domain and they are simply better and more efficient tools for that domain."

Grep the Web

Slides and talks from the recent Hadoop Summit are now available. Some of the more interesting ones is Facebook's Hive, Amazon's GrepTheWeb, IBM's JAQL and Yahoo's just about everything else.

Engrich

The ability for a foreigner to speak just enough English in order to swindle stupid Westerners.

Update from WWW2008

The HCLS workshop was very good. I especially enjoyed Mark Wilkinson's talk about BioMody 2.0 (very Larry Lessig-esque) and Chris Baker's. There's some definite interest from a number of people about my talk too.

The keynote of the first day was from the Vice President of Engineering at Google, Kai-Fu Lee. In my talk I said that IBM had noted that scale-out architecture gives you a 4 times performance benefit for the same cost. He said that Google gets around 33 times or more from a scale-out architecture. The whole cloud thing is really interesting in that it's not only about better value but about doing things you just can't do with more traditional computing architectures. The number of people I've overheard saying that they haven't been able to get their email working because they're using some sort of client/server architecture is amazing. I mean what's to get working these days when you can just use GMail?

The SPARQL BOF was interesting as well (Eric took notes). The time frame seems to be around 2009 before they get started on SPARQL the next generation. What sticks out in my mind is the discussion around free text searching - adding something like Lucene. There was also aggregates, negation, starting from a blank node in a SPARQL query and transitive and following owl:sameAs. I was pretty familiar with all of these so it was interesting just to listen for a change. So with both aggregates and free text you are creating a new variable. Lucene gives you a score back and I remember in Kowari we had that information but I don't think it was ever visible in a variable (maybe I'm wrong I don't really remember). It would be nice to be able to bind new variables somehow from things in the WHERE clause for this - and that would also allow you to filter out based on COUNTS greater than some value (without having a HAVING clause) or documents that match your Lucene query greater than a certain value. Being able to do transitive relationships just on a subset of the subclass relationship (like only subclasses of mammals not infer the whole tree of life) seemed to have been met with some reluctance. I really didn't understand this but it seemed to be around that it was the store's responsibility to control this and not up to the user to specify.

The other thing that was mentioned was transactions. It seems that transactions probably won't be part of SPARQL due to the nature of distributed transactions across the Web.

There was one paper on the first day that really stood out. I don't know what it is about logicians giving talks but they are generally really appealing to me. It was "Structured Objects in OWL: Representation and Reasoning" presented by Bernardo Grau. It seems to take the structural parts of an OWL ontology and creates a graph to represent it. This prevents DL reasoning of an infinite tree and creates a bounded graph. This is cool for biology - the make up for a cell for example but it also speeds up reasoning and allows errors to be found.

The other interesting part was the linked data area. I was a bit concerned that it was going to create a read only Semantic Web. A lot of the work, such as DBpedia that converts Wikipedia to RDF, seems a bit odd to me as you can only edit the Semantic Web indirectly through documents. But in the Linked Data Workshop a paper was presented called "Tabulator Redux: Browsing and Writing Linked Data" which of course adds write capabilities. I spoke to Chris Bizer (who gave a talk on how the linked data project now has ~2 billion triples) about whether you could edit DBpedia this way and he said probably not yet. That's going to be interesting to see where it goes.

I am just going off memory rather than notes. So I'll probably flesh this out a bit more later.

Big Web Table

I thought I read someone ports Google's AppEngine to use HBase. Good idea but not quite. "Announcing A BigTable Web Service": "I then came up with the crazy idea to offer BigTable as a web service using App Engine. It would be an infinitely scalable database running in Google's datacenters. I spent my weekend learning Python and hacking together an implementation. Now I'm happy to present the BigTable Web Service. It models the API of Hbase—a BigTable clone. Now you can have simulated BigTable running atop App Engine, which itself provides an abstraction on top of the real BigTable."

What it actually does is use HBase's Thrift API on top of Google's BigTable or as they say BigTable as a Web Service (a RESTful one).

Mario's Jazz Bar

Just a random thing to share - there seems to be quite a lot of competition playing Super Mario Galaxy songs on YouTube. A fairly recent one is a Jazz Interpretation of the Observatory theme (there's also a guitar version or accordian one). I still quite like the original orchestral version of Gusty Garden Galaxy Theme and it's version on piano. Koji Kondo is also good to look up on YouTube from time to time too.

What Women Want: Pairing

It's not the first time I've read an article about the continuing decline of women in IT, "Where Did All the Girl Geeks Go?" continues to note the slide:

"There's a perception that being a computer science major leads to a job as a programmer and you sit in a cubicle where you type 12 hours a day and have no interactions with other people," Block said.

Yusupova noted that even if pure programming jobs are outsourced, opportunities still remain within a company for people to bridge the relationship between the outsourced IT vendors and the business side.

"These roles would probably be ideal for women who prefer to be in communication-focused roles, if they know computer science and can communicate to all parties involved," Nelly Yusupova, chief technology officer of Webgrrls International, a networking organization.

There's was a talk given recently, at a local XP group, that lead to a discussion on the benefits of things like pair programming (see "Pair Programming").

I see pair programming and other means to improve interactions between developers not only essential for better code and a better project but also as a way to improve the IT industry generally and to expand its appeal especially to younger people and women. The idea being that certain people work better in a participatory manner rather than being told what to do.

This is pretty much what an article a couple of year ago suggested called "Debunking the Nerd Stereotype with Pair Programming" (or as PDF):

Jamie wants to be a software engineer. She enjoyed her programming and science classes in high school and wants to combine her interest in both disciplines to help society through biomedical applications. Since she started college, it seems that her life has been centered on time consuming programming classes. In those classes, her professors insist that she work alone—some professors expressly forbid even discussing assignments with fellow classmates. Before entering college, Jamie was aware of the stereotypical view that programmers work long hours by themselves. Based on her college experience, now she knows it’s more than just a stereotype—it’s true. Perhaps she should forget programming. She likes the friends she’s met in her biology lab group—maybe biology would be a better major.

Having been working in bioinformatics for over a year it's startling the number of women in this area compared to IT. It seems basically 50/50 in what is essentially an application of information technology. They still write code, they still develop large applications and so on. Why does it drop to 1 in 20 or worse in IT? It does seem that in bioinformatics you are expected to work in groups and teams, they are forever interacting with each other - it seems a brilliant environment as far as productivity is concerned.

And it's not just IT or biology but it seems that there is a general benefit from greater interaction, more pairing and the like generally improves performance:

The success rate of underrepresented minorities in science courses has been shown to be dramatically improved by shifting the learning paradigm from individual study to one that capitalizes on group processes, such as student work groups and student-student tutoring.

From an IT perspective pairing doesn't only improve the quality of the software it also improves your abilities as an individual programming as well, as has been demonstrated where pair programming has been used in IT courses and the results of students in exams improved (see "Pair Programming Improves Student Retention, Confidence, and Program Quality").

I re-read "All I Really Need to Know about Pair Programming I Learned In Kindergarten" which still holds up quite well as a set of rationales behind pair programming and NCSU's Pair Learning has lots of papers related pairing, learning and making IT more attractive to more discovery based system.

Update: Finally found a public version of the nerd article.

hashCode and equals for Blank Nodes

You don't need node ids. Most, if not all RDF triple stores take a Literal, URI Reference or Blank Node and generate a node id. Sometimes it's a hash or UUID, sometimes it's from a node pool or value store but you don't really need it. As an aside, in a distributed store you could even do the blocks of ids trick which people have done in SQL databases but I haven't seen that done for RDF yet.

When you do operations, like joins, in Java or Ruby or some other language you rely on hash codes to generate different values, if they're the same then you call equals.

What if you don't have a node pool?

It's easy to do for what I like to call globally addressable values - URI References and Literals - no matter where you are, these methods return the same results from their hash code or equals. Not so with Blanks Nodes, which are tied to the context of an RDF graph.

One solution is to ban blank nodes - they're pains to parse, query and store. But I actually like blank nodes. There good at representing things where you don't want to confuse it with something that might actually be a URI to dereference.

The idea we've been working on with our high-falutin' scale-out MapReduce blah blah is really just coming up with sensible implementations of the hashCode and equals methods for blank nodes. There is previous work done in distributing blank nodes across graphs, the one that I'm most familiar with is RDF Molecules. But they didn't really quite cut it as far as hash codes and equals are concerned and that's basically what I'm presenting next week in China. The hash code is basically the head triple and the equals is the minimal context, sub-graph for a given blank node.

There's a lot more to say, as I've had to find something to talk about for the whole 15 minutes.

My (Continued) SPARQL Debacle

One of the reasons I started this blog was to record my current thoughts at a particular time. With this in mind, I should track my recent comments about SPARQL and the empty graph pattern and further rehashing of it.

I made a few mistakes during the discussion and spent well over a week in discussion and maybe a week prior to asking the question thinking about it and much time thereafter just thinking about summarizing it.

SPARQL is an algebra that is not consistent (isomorphic) with what I think of as set/relational/bag algebras (even though it appeared at one stage this was considered). The reason is that identities I believe hold in these algebras don't for SPARQL.

The set/relational/bag algebra identities are:
* A + 0 = A * U = A
* A + U = U
* A * 0 = 0

Where + is UNION, * is INTERSECTION, A is any set, 0 is the empty set and U is the universal set. The second one is expressible and does work in SPARQL. The first one isn't expressible in SPARQL. The last two don't hold.

You can derive the last two identities from the first two as long as you have compatible definitions for things like inverse (or complement). When Date creates the algebra for bags he spends most of his time coming up with a reasonable definition for the complement of a bag which seems to be more like difference. In my interpretation 1/T/U is the relational TABLE_DEE and 0/F/empty set is TABLE_DUM. I thought this is quite clear but it appears even this is up for interpretation.

Prior to Date's latest book, I had a bunch of his writings which I used to create identities for OPTIONAL, JOIN and UNION. I struggled a while back to see whether they were compatible with SPARQL, which I eventually decided that they were compatible, it ends up that they are not - because the identities don't hold. I think as long as you don't ask these questions then it still returns the right answer and you could create special cases for SPARQL's empty graph pattern but I'm just not that confident anymore. SPARQL is more like an algebra of numbers than of sets or bags.

Reflecting on this, I was striving for a consistency that just wasn't there and if I squint hard enough I can see how the SPARQL algebra by itself makes sense.

There is still some behavior, even within the SPARQL specification, that appears to be really bad like "SELECT ?x WHERE { ?s ?p ?o }" being a valid query (it returns a number of unbounds to ?x for however many triples there are in the graph). This is quite different to SQL or relational PROJECT. It's also weird that the SPARQL specification is different to the Perez papers about SPARQL - evaluation is done at a grammatical level. UNION also differs as it's defined as multiset union not set union even though OPTIONAL is made up of set union not multiset union. Actually, I'm still not sure if UNION is multiset union because in the implementations I've seen the order is important (that is {} UNION {} UNION { ?s ?p ?o } is different to {} UNION { ?s ?p ?o } UNION {} and { ?s ?p ?o } UNION {} UNION {}) but I guess that's because of the grammatical evaluation.

It does put any further work on JRDF's SPARQL implementation in a bad position. I can keep calling it SPARQL but know that it's not following the standard or rename it (currently I'm thinking URQL) but the whole point of bothering seems to be questionable. The ironic thing is that it could pass all the SPARQL tests even though I know it's not compatible. In other work that I've been doing, I've been interested in SPARQL as the Unix pipes for RDF and blank node round tripping but SPARQL doesn't work there either. Blank node round tripping is where you take the result of one SPARQL query that includes a blank node and put it into a second.

Sometimes you come away from asking a question feeling validated or smarter and sometimes not. This time it's definitely not - I no longer feel confident talking about SPARQL or relational algebra anymore.

Microsoft LINQs Data

Microsoft and "Research-Output" Repositories

Our goal is to abstract the use of underlying technologies and provide an easy-to-use development model, based on .NET and LINQ, for building repositories on top of robust technologies.

The platform has a "semantic computing" flavor. The concepts of "resource" and "relationship" are first-class citizens in our platform API. We do offer a number of "research-output"-related entities for those who want to use them (e.g. "technical report", "thesis", "book", "software download", "data", etc.), all of which inherit from "resource". However, new entities can be introduced into the system (even programmatically) while the existing ones can be further extended through the addition of properties.

This means, obviously, that arbitrary relationships between resources can be established. Our platform comes with a number of "known" predicates (e.g. "added by", "authored by", "cites", etc.) but it is extensible to accommodate any new predicates developers want to introduce. Furthermore, we do not interpret the semantics of the relationships; we let applications define how to reason about them.

The concept of a "relationship" may make many think that we are building a triple-store, perhaps even speculate that we are using one. While we do store tuples, we have opted for a hybrid approach between a fully-blown relational schema and a triple-store. Our thesis is that by sitting in the middle of the "triple store <–> relational schema" spectrum, we will be able to stay flexible enough without impacting performance.

At the Open Repositories 2008 conference, we will formally unveil our work in advance of its official release and initiate interactions/exchanges with the DSpace, EPrints, Fedora, and other players in the repository community. This is crucial to us because—like every other project our group undertakes—we are intensely focused on interoperability.

Maybe Microsoft and Yahoo! have more in common than previously thought.

Via, Microsoft set to launch Semantic Web light. I previously looked around for LINQ tools for RDF.

Save Ontologies from the Ontologists

I presented a talk at InterOntology08 last week (list of slides presented). It was only 15 minutes so there wasn't room for much content. What I think is the most important slide was number 11 about how the BioMANTA project is attempting to produce ontologies as an agile, engineering artefact that are verified in reality (due to experiments being performed, provenance tracked and data analysis on the quality of the provenance to filter out irrelevant or incorrect data).

There were some good things to come out if it. Thinking about how to describe to other people problems with ontologies in terms of inconsistencies - what will be inferred that contradicts your ontology - was very useful. The work done by Werner Ceusters, Nicola Gurino and Yu Lin were the most close to our work. One of the speakers gave what I think as a succinct description of the difference between top down vs. bottom ontology development: "what to expect" vs. "what to extract". I also met a lot of great people who I hope to meet again and Japan was very cool.

Easily the best, in terms of the most thought provoking, was Barry Smith's, "The Evaluation of Ontologies: Editorial Review vs Democratic Ranking". This discussed the work of the Gene Ontology and the OBO Foundry. He cited the Gene Ontology as the most useful and most used ontology which has been developed using a top down process. It allows comparable data to be produced, it removes data silos and he compared it to creation of standard measures (metric system). He said that in order to achieve this standardisation you need editorial committees. An ontology becomes part of the peer review, journal process. He introduced the OBO Foundry which has many principles such as being open, has a formal language, collaborative, orthogonal components, versioned, well documented and must have data before it can be accepted.

The alternative view he offered was attributed to Mark Musen. It's a bottom up, annotation of ontologies and many of the slides were taken from a previous talk. Mark believes that ontologies are still a cottage industry and that it is often difficult to ascertain the quality of an ontology just by inspection. He said it is also true that we may wish to use parts of ontologies even if they are not well designed. He questions whether a top down approach can scale. He is developing BioPortal which offers a way to upload and rate various ontologies. The key question about BioPortal is whether it will generate enough interest to reach a critical mass of reviews.

I had many problems with this talk. Firstly, the way it was characterised as one vs. the other - why can't they both work? What stops peer reviews of popular ontologies or getting popular ratings of peer reviewed ontologies. Barry mentioned that a selection approach works for refrigerators (where peer review designs the function of the refrigerator and colour is selected by the masses) but questioned whether this should work for science. This is an obviously negative view of what mass selection can do - we choose representatives in a democracy or successful products in a capitalist market, surely these are very important things that are left to the masses. Are ontologies any less than these things?

Beyond that, both of these methods seem to suggest a certain centralisation. Doesn't this encourage gatekeepers, people holding onto power, hasn't the web (governments, capitalism, science etc.) shown that decentralisation is better? I see science as a competition of ideas, the best model is chosen over many possible ones that best fits existing data and predicts new observations.

One of the OBO Foundary principles is that you can't reuse an ontology. That is, if you're outside the OBO Foundry and you make a change you can't redistribute or use the same identifiers. This just seems wrong. I must be misunderstanding this part, because it is supported by people who I would expect to support the idea of reusing ideas and, most importantly, sharing them with others.

Many of these arguments seems to be around whether an ontology is attempting to create or represent reality or if its an engineering artefact. I see it as a bit of both but its primary utility, I'd suggest, is as an engineering artefact. It represents a (hopefully working) system.

A simple example is our "fixing" of BioPAX. BioPAX uses string literals for certain properties and this prevents them being used as subjects in RDF. I would like to link, maybe dereference them and do other cool things with them that you can only do with URIs. So I'd like to make a change now, get something working and distribute my software with these changes.

I do think that ontologies should be well documented but documentation can be a barrier when you want to change something, try it out, make more changes, try it out again - the documentation is potentially going to be missing or wrong. The whole process seems to be trying to do too much upfront - which is terrible for the few, overworked ontologist that there are.

I don't want to wait while my ontology gets peer reviewed necessarily - the chances of the right person finding a mistake really doesn't sit with a committee or voting process - I'd like it to include everyone. I'd like to do it cheaply both in time and money; if for no other reason than to see whether it works well. If it doesn't work then it's not a big deal I can just change it back. It seems that if this was part of a big process then it would be less likely to happen.

Both methods also lack verification (or at least it wasn't discussed). There's nothing to say that a bunch of people in the OBO Foundary or a voting process will necessarily achieve certain modelling objectives - something that is right for me or for everyone. Like most systems, ontologies will have contradictory requirements such as flexibility and completeness or security and privacy - there really isn't one true answer. I'd prefer a process that quickly adapts to changing requirements which can then be verified.

PURLs for GO

I think these have been published before but I only just noticed the use of PURLs for references in the Gene Ontology. For example: http://purl.org/obo/owl/GO#GO_0008150 (the gene is not dereferenceable but then that seems okay for current purposes). This is following Recipe 1a from the "Best Practice Recipes for Publishing RDF Vocabularies", although the file sizes are probably too big (as suggested in "Serving Static RDF Files"). It does seem inconsistent with the Banff Manifesto which suggests URLs more like http://purl.org/bm/go:0008150. I know about slashes and hashes but I'm not sure about colons.

Algebra A and SPARQL

I've been reading, "Logic and Databases: The Roots of Relational Theory" and more importantly the chapter 10 which is about "Why is it called a Relational Algebra?". He defines what an algebra is such as identities, idempotence, absorption and so on with respect to Algebra A. I first came across Algebra A in the 3rd Manifesto which is an untyped relational algebra that defines a relationally complete system in about three operations: REMOVE, NOR or NAND and TCLOSE (transitive closure). I say, "about three" because I'm not sure TCLOSE is part of a relationally complete system and NOR and NAND are made up of an untyped OR or AND plus NOT. It also gets rid of WHERE, EXTEND and SUMMARIZE (which can be used for aggregate functions) by creating "relational operators" which are special relations that perform an operation (like COUNT).

Anyway, one of the more interesting points is that on page 260-261 of "Logic and Databases" he talks about identities such as: A + 0 = A * U = A, A + U = U and A * 0 = 0. Where A is any relation, 0 is the empty relation (DUM) and U is the universal relation (DEE). These match the tables I created for JOIN and UNION for SPARQL - and likewise I think are correct for OPTIONAL.

There is also a chapter on the closed world assumption and why Date dislikes the open world assumption which I'm still trying to digest. It seems that to get around 3VL Date uses strings - which seems like a massive hack.

It also occurred to me when reading this that SPARQL and query languages in general are non-monotonic - that is as you add more information the results you get from a query can be different - which is different to RDF. It made me wonder what a monotonic query language would look like but not for too long.

Tissue Parade

This is just a quick note to let people who have expressed interest before in BioMANTA that the web site is now pretty much up-to-date with the latest papers and presentations (except for the InterOntology08 presentation in Japan next week). If we've sneezed and there was a Powerpoint slide it's there.

Do You Have an Internet Sized Problem?

Yahoo! Launches World's Largest Hadoop Production Application. This is mainly about Hadoop reaching a certain level of maturity.

The Webmap build starts with every Web page crawled by Yahoo! and produces a database of all known Web pages and sites on the internet and a vast array of data about every page and site. This derived data feeds the Machine Learned Ranking algorithms at the heart of Yahoo! Search.

Some Webmap size data:

* Number of links between pages in the index: roughly 1 trillion links
* Size of output: over 300 TB, compressed!
* Number of cores used to run a single Map-Reduce job: over 10,000
* Raw disk used in the production cluster: over 5 Petabytes

As described in the video, Webmap is the directed graph of the web and store the aggregate the metadata about the links. A lot of the code sounds like it is written in C++ (which is why there's the pipes API in Hadoop). According to the Hadoop mailing list, this means that Webmap is roughly equal to Google's scale.

Sorry

Maybe I'm in leftist, socialist heaven here but it does seem that most people were following the apology to the stolen generations. People were sitting in their cars listening to it rather than going to work. I liked the focus on how saying sorry is not about making you feel better it's about the person (or people in this case) you're saying sorry to and that you have to take the good with the bad about your country and make amends. There is also a sense that some sort of action will take place; with both sides of parliament working together. A lot of the Howard policies were designed as wedges to divide, so while I'm very skeptical, I'm a little bit hopeful that some inclusive politics will occur.

There were a couple of moments I thought worth mentioning from Rudd's speech:

After living in Alice Springs for a "few years", government policy changed and the young girl was handed over to the missions.

"The kids were simply told to line up in three lines ... those on the left were told they had become Catholics, those in the middle, Methodist and those on the right, Church of England," Mr Rudd said.

"That's how the complex questions of post-reformation theology were resolved in the Australian outback in the 1930s.

"It was as crude as that."

Mr Rudd said should there still be doubts, the historical record showed that between 1910 and 1970, between 10 and 30 per cent of indigenous children were forcibly taken from their mothers and fathers.

"As a result up to 50,000 children were forcibly taken from their families," he said.

Mr Rudd said one of the most notorious examples of this approach came from the Northern Territory Protector of Natives, who had stated: "Generally by the fifth and invariably by the sixth generation all native characteristics of the Australian Aborigine are eradicated. The problem of our half castes... will quickly be eliminated by the complete disappearance of the black race and the swift submergence of their progeny in the white."

"The 1970s is not exactly a point in remote antiquity," he said.

"There are still serving members in this parliament who were first elected to this place in the early 1970s.

It's been a long time, I remember Paul Keating's Redfern Park speech which was voted the number 3 speech after Martin Luther King and Jesus.

Concept Extractor

Last year ClearForest was bought by Reuters and Tim O'Reilly has covered the story of how the CEO of Reuters sees a Semantic future. This is one of those companies that I'd hope would become semantically enabled way back when. It would be good for other text mining companies like Autonomy and Inxight to jump on board too. The free web service provided "categorizes and links your document with entities (people, places, organizations, etc.), facts (person ‘x’ works for company ‘y’), and events (person ‘z’ was appointed chairman of company ‘y’ on date ‘x’). The metadata results are stored centrally and returned to you as industry-standard RDF constructs accompanied by a Globally Unique Identifier (GUID). Using the Calais GUID, any downstream consumer is able to retrieve this metadata via a simple call to Calais".

No Crappy Wrappers

Charles Petrie has an article called, "Is Semantic Web Technology Taking the Wrong Turn?". The author suggests that the current direction in Semantic Web development is leading it towards irrelevance. He notes its requirements and its attempt to provide ways to simplify and speed up tasks, such as integration, by an order of magnitude over existing technologies.

He sees a problem with how Semantic Web Technologies (SWTs) have typically been applied by adding layers alongside existing ones. This just increases "the number of interfaces and mediations required". Furthermore, most publications using SWTs talk about homogeneous environments - languages, ontologies, definitions are all constrained and any differences avoided. The other mistake highlighted is that the work has been divided into the usual architectural layers (persistence, processes, UI, etc) which has lead to each of these layers having their own Semantic Web layer added - creating "a disaster for software architects and engineers who must use the results from several communities in building software applications that are hosted and interconnected".

There are two obvious ways you could attack this argument. The first is that these types of constraints have been applied because of the immaturity of the underlying systems. It's hard to develop an integrated system in one go - dividing up these problems into their layers is obviously one way to make progress. There is also obvious infrastructure missing, not just better triple stores, but also ways to make sure you can reuse ontologies and processes. The other is that there are many examples of aligning ontologies, reusing them, and merging concepts from them; but it too is still in its infancy. Areas like data mining, that the Semantic Web could leverage, lacks mainstream use as well.

Straight ahead from here leads to more SWT languages, hard-to-integrate ontologies, and technology components such as libraries, RDF databases, and logic reasoners. Those who build real-world applications will have to integrate all those elements to use them holistically, thus leaving the integration problem unresolved. As this approach increases the effort required in every part of the software engineering life cycle, chances are that developers will adopt the SWT only for very specific areas and solutions, rather than for general use across all domains in which computing is applied.

He offers a possible solution in addressing the data and process heterogeneity by not restricting SWT to the edges of systems but for it to be applied throughout systems.

Rather than looking at SWT as interface-wrapping technology, it seems appropriate to make it the foundation for all aspects of information technology and scientific computing. In concrete terms, one way to eliminate mediations when crossing layers is to ensure that data objects are encoded in a single format (such as RDF) and not mapped between layers but rather handed over from layer to layer without change. This, in turn, would challenge the various technologies used for implementing these layers to become totally SWT aware.

I think that development goes through cycles of integration and separation but I do agree that if the Semantic Web is just a technology of wrappers it will fail.

Update: Much along similar lines is an article about Dieter Fensel, "Are Semantic Researchers Missing the Big Picture?", he says:

...we do a lot on research of apply[ing] in semantics to all aspects of Enterprise Application Integration where you integrate data, processes, and services (and not only web pages)...

Is the industry neglecting the greater overall goals of scalability for interoperability?

“No,” writes Fensel. “I think they are aware of [it]. For example, Michael Broodie, Scientific Director at Verizon, estimates that world wide around 1 trillion dollars are spent per annumn on application integration. The semantic web community (and not the industry) is mostly ignoring this area.”

I do like the cycle though, we've gone from an initial SEMANTIC web and criticism, to semantic WEB and now this criticism and back to highlighting semantic again.

50% Less Code or Your Money Back

Less that two weeks ago 0.5.3 of JRDF is released and now it's 0.5.4's turn.

This is mainly driven by making the fix to the bug in the btree (or Sesame's version) available.

It does however mean that the changes to the Resource API are made available more quickly. I remember looking at Jena's Resource and thinking it was bloated, confusing and rather poorly thought out from an efficiency point of view (holding onto all Graph and associated objects) but now I finally understand the positive effects it has on the code and I like it. JRDF's Resource sits on top of an RDF graph and automatically performs conversion from Java objects like URIs (which become URIReferences) and insert or removes them from the graph. It too has many methods that do the same thing but with different types. It's better than the recent changes to TripleFactory as it allows blank nodes too. JRDF's Graph implementation is no longer the heavy weight object it once was, it references the indexes and nodepool but no longer has any real control over them (this meant removing serialization).

Code becomes a lot smaller too, for example (the create call is URI.create):

Resource supplier = elementFactory.createResource();
supplier.addValue(create("urn:supplier"), "S1");
supplier.addValue(create("urn:name"), create("urn:Smith"));
supplier.addValue(create("urn:status"), 20);
supplier.addValue(create("urn:city"), "London", XSD.STRING);


Creates the triples:

_:1 urn:sno "sno"
_:1 urn:name urn:Smith
_:1 urn:status "20"^^xsd:int
_:1 urn:city "London"^^xsd:string


It used to be something like:

Resource supplier = elementFactory.createResource();
URIReference supplierPred = elementFactory.
createURIReference(create("urn:supplier"));
graph.add(supplier, supplierPred, "S1");
...


Scaling MapReduce

Before I started down using Hadoop I had to be roughly sure that it would scale. One of the pieces of evidence that convinced me was done by IBM in "Scalability of the Nutch Search Engine", they performed tests using real world systems and also modeled the architecture to verify it:

We observe...that the throughput, for a fixed data set size per back-end, increases with the number of back-ends.

We observe a generally good agreement between measurement and prediction for small number of back-end servers. The response time is essentially flat with the number of back-end servers, up to 500, 1000, and 2000 servers for data set sizes (per server) of 10 GB, 20 GB, and 40 GB respectively. The system is more scalable for larger data set
sizes because the work per back-end server per query increases.

They are able to service 1000+ nodes with a single server.

Kids Today

You tell them that there used to be monkey shaped pixels and they wouldn't believe you.

Nodes by Type

JRDF 0.5.3 released. This has a couple of bug fixes and features added. The find with no constraints (find(ANY_SUBJECT_NODE, ANY_PREDICATE_NODE, ANY_OBJECT_NODE)) has always been fast with the on disk version but not if it was constrained in anyway (a fixed subject for instance). This was because the wrong method on the btree was being called. This is a big improvement in performance obviously.

A cleanup on the Graph interface also occurred. There were all these methods like findUniquePredicates, getBlankNodes and getResources. There is now findNodes(NodeType) where NodeType is any of the positional (SPO) or value (Resource, URIRef, bnode, Literal) node types. This was able to be done because JRDF has a NodeTypePool where all nodes are in maps by their type. It is missing some types being supported but that should be done next release.

The other thing that was very annoying and now fixed was that a Literal or URI Reference had to be added via the GraphElementFactory before you could add it to the Graph. This only makes sense for blank nodes - other node types can just be re-localized (get their internal node id) each time. Handy for the amount of graph copying we have done lately.

There were some other boring changes like a bug fix to the RDF/XML parser and TripleImpl now being null proof.

Much more to do - haven't had a chance to cleanup the Molecule implementation or add the persistence support.

Solid State Drives

I was wondering what the effect would be on data structures and algorithms given the new features of SSD (solid state drives) - that is, drives with ten times the seek performance and improved read speeds. Desktop hard drives are about the same (18% slower) or beat SSDs in a straight line, sequential access, especially writes. In the short term, it looks like solid state drives biggest impact is likely to be in providing laptop drives the same performance as desktop ones.

A secondary short term impact may well be in providing another level of storage between spinning hard disks and other caches. This is mentioned in, "The five-minute rule twenty years later, and how flash memory changes the rules".

The name of their rule refers to the break-even interval between accesses. If a record (or page) is accessed more often, it should be kept in memory; otherwise, it should remain on disk and read when needed.

Not surprisingly, the optimal page size for B-tree indexes on modern high-bandwidth disks is much larger than traditional database systems have employed. The access time dominates for all small page sizes, such that additional byte transfer and thus additional utility are almost free. B-tree nodes of 256 KB are very near optimal...a traditional rotating hard disk, Table 3 indicates 337 seconds or just over 5 minutes.

Due to the lack of mechanical seeking and rotation, the transfer time dominates even for small pages. The optimal page size for B-trees on flash memory is 2 KB, much smaller than for traditional disk drives. In Table 3, the break-even interval for pages of 4 KB is 351 seconds.

Using O’Neil’s SB-trees, extents of 256 KB are the units of transfer between flash memory and disk, whereas pages of 4 KB are the unit of transfer between RAM and flash memory.

Mentions this paper, SB-Tree : An Index-Sequential Structure for High-Performance Sequential Access.

Via, Flash Memory and Databases.

Years to Days

Dr J Craig Venter – A DNA-Driven World (video)

For the past 15 years at ever faster rates we have been digitising biology. By that I mean going from the analog world of biology through DNA sequencing into the digital world of the computer. I also refer to this as reading the genetic code. The human genome is perhaps the best example of digitising biology. Our computer databases are growing faster per day then during the first 10 years of DNA sequencing. The databases have been filling even faster with the results of our global ocean sequencing project. As a result we have now over 10 million genes in the public databases, the majority of which have been contributed by my teams.

Instead of evolution happening only due to random mutations that survived selective pressure, we can see how by adding chromosomes to or exchanged between species, that thousands of changes could happen in an instant.

Now they can happen not just by random chance but by deliberate human design and selection. Human thought and design and specific selection is now replacing Darwinian evolution.

The biggest question in my mind is the one of scale. Last year we consumed more than 83 million barrels of oil per day or 30 billion barrels during the year. In addition we used over 3 billion tons of coal. These are mind boggling numbers and the only way that I can see replacing oil and coal is through a widely distributed system. If there were one million bio-refineries around the globe each one would still need to produce 17,000 liters per day.

He also talks about the rise of fundamentalism, that 25% of people in the US don't know that the Earth revolves around the Sun, half think people and dinosaurs co-existed and that 58% don't know how to calculate a 10% tip.

Who Killed Functional Programming?

MapReduce: A major step backwards. Two good responses, "The Great MapReduce Debate" and "Relational Database Experts Jump The MapReduce Shark":

MapReduce is not a data storage or management system — it’s an algorithmic technique for the distributed processing of large amounts of data.

MapReduce has the same relationship to RDBMSs as my motorcycle has to a snowplow — it’s a step backwards in snowplow technology if you look at it that way.

I don’t think the authors understand distributed processing and distributed file systems when they think reduce must rely on FTP. The Wikipedia article says “each [reduce] node is expected to report back periodically with completed work and status updates,” pretty much the opposite of the “pull” DeWitt and Stonebraker criticize.

It's a fairly specious argument that more features means better and the comparisons are not equivalent - it would be better compare BigTable or HBase with databases not MapReduce - and that'd be silly because these guys are column database proponents which is (roughly) the same as HBase/BigTable (they do say it lacks views but the idea used by Google is copying rather mutating). An example of using MapReduce to do indexing, is Nutch. The distributed filesystem is important and it should be clear by now, you'd hope, that storing the data is not the problem, processing it is.

These types of comparisons, "X does something Y doesn't therefore X is better", reminds me of "Who Killed the Electric Car?". One of the reasons the EV1 was said to have failed was because "lacking an engine, it saves the driver the cost of replacement parts, motor oil, filters, and spark plugs" (and more) - all those moving parts, that had to maintained were gone. How can you change gears if you don't have a gearbox?

Does that remind you of a database or other piece of software you use? Isn't the software industry just like the car industry? There appears to be market forces ensuring that things stay complicated and expensive to maintain.

Meanwhile, Google is getting on with it by providing services for terabytes of data (via Wired), "Building on the company's acquisition of the data visualization technology, Trendalyzer, from the oft-lauded, TED presenting Gapminder team, Google will also be offering algorithms for the examination and probing of the information. The new site will have YouTube-style annotating and commenting features."

Update: It's been noted that bet on cheap and rickety if it's 10 times better (also mentions hypertable) and that they're comments are not even wrong.

Update 2: Part of my rant was going to be the whole Apple designs things with the fewest stuff in them - that's what makes them the leader what they leave out. It's the same gist that I read at "Heavier than Air".

Fusion Beats Transistor

Why fusion? Answered fairly well in, Big machines for big questions. The presenter says that fusion has been outstripping CPU development, the triple product in fusion generation doubles every 1.8 years vs 2 years. They talk about ITER being the first fusion plant to generate more energy than put in. Worth the price of admission to the 21st century alone.

Ricky Gervais Explaining what's Wrong with Evolution

Link.

It's Fun to Pun

Punning was one thing that stumped me about OWL 1.1 last time I looked - mainly because I got confused between it and OWL Full. We did similar things for ontologies such as the Cell Type ontology (although the names were not the same in order to avoid moving out of DL land).

Recently, there's been a nice thread on "Xtreme Punning". I quite like Pat Hayes description of the situation:

Conventions which everyone has to agree to use: which will never happen. Names are cheap, but agreement on names is not cheap, so the fewer names we can get away with the better. The datatype example in my earlier email is a good example, IMO: all three uses are quite natural, and I wouldn't want to have to remember to distinguish xsd:number from rdfs:numberMapping from owl:numberThing.

Names can be freely used in ANY logical syntactic role, without ANY restrictions. EVERY occurrence of a name denotes the same thing. ALL names denote both a class and a property (and in CL a function; and in IKL, a proposition. No doubt other ideas can be incorporated.) ANYTHING can be an instance. Hence, any two legal sentences
(ontologies) can be simply concatenated and the result is still legal (no checking to be done when merging graphs) and meaningful. And it all works very simply, with a simple uniform semantics, which one can quickly get used to.

RDF in a...database?

This is has many similiarities in a recent conversion about the relational model and RDF, comes "Relational databases for storing and querying RDF":

Our work in this area, "Scalable Semantic Web Data Management Using Vertical Partitioning," appeared in the VLDB Conference in Vienna in September. It showed that using a column-oriented database, along with this property representation, allows us to overcome the overhead of representing NULLs, while providing two orders of magnitude better performance than the naive triples representation. This is particularly true when processing queries that must access many triples during execution (e.g., computing the number of books grouped by subject area or institution.) Of course, there is a fair amount of subtlety to getting good performance out of such a representation. Have a look at our conference paper for the details!

This paper and others was mentioned previously. This view, of property tables in column databases, is very similar to untyped relations.