2nd Web conference

This morning Christof and Horst we had the 2nd web conference. This time the Polycomm does not work well. Now and then we hear high frequency noisy. Horst said it’s because he is adjusting his earphone.

Got new work to do. Horst will arrive this Friday. He asked if I want we can meet up on Sunday. I would rather not. hahaha.

We further discussed the requirements for doing the assignment using AIS. However, I think the bottleneck is on my side. I have to make it very clear what I want to accomplish.

Horst thinks in a longer term perspective about what features AIS should incorporate. But Christof tried to make the task as specific as possible so that I can use the existing features of AIS and finish the case study before Oct.

Something good to have

• The current AIS API only support tree taxonomy, no multiple parents are allowed.
• Association cannot has attributes, only types.
• Association types uses a different set of terms from the Info Item types.
• Only one workspace is active in AIS context.
• Info Items are only associated with Info Items, no association can be made between an Info Item and an InfoItemSet. (Horst mentioned Prototo supports this.

Horst mentioned Yahoo Pipes to model the process.

 

Web Conference

Last time when I talked to Horst, we used Skype. Strangely the quality is not good. We then used the telephone line.

This time Christof and me found a conference room and we used Polycomm and NetMeeting. Ploycomm’s sound quality is amazing! No wonder Fry’s tags it for over 500 bucks. It has been ages I used NetMeeting. But it still works. We used Desktop sharing feature. Even though NetMeeting doesn’t have those fancy internet messenger features, but it can share Desktop or a specific Windows application such as Firefox or Notepad.

Meeting takes 1.5 hours. Not lengthy. German are efficient, especially Christof. Every discussion with him leads to action. No miss after this meeting too.

The meeting is about to evaluate the use case I proposed to work on top of AIS. I first introduced the genomics use case that Dr. Butler and I found it interesting and practical to biologists. It took them for awhile to recall their Chemistry and Biology subjects. I know! Me too. I spent the whole Labor day weekend working on getting to the terms. I don’t expect them to know it so much faster in a meeting! But it is an interesting use case. We found some good fit to AIS, but some features need to be added into AIS. For now we try to work around the needed features by fully making use of the current API.

We will meet tomorrow. Horst is coming down to Palo Alto next Monday. We can have a live conference by then!

As for them, they are more interested in the applicability of AIS API. For me I am interested in how this real world problem can be modeled and automated.

 

Taxonomy Importer

After two days, I finally finished the implementation of taxonomy importer. My task is to import a taxonomy in XML format to an internal tree structure. Each taxonomy term will be a tree node. The AIS API then can call on the terms for further operation.

I pick JDOM as the XML parser. It seems simple to use. And have been proven by many developers. Even Horst has used it in AIS API.

I spent some time thinking whether I shall rely on JDOM to perform tree operation each time I need to retrieve the taxonomy nodes as JDom internally parses the whole XML document into a tree structure. However, it is not a clean design. What if I need to perform more than one operation on this tree structure? And the DOM tree built by  JDom has many nodes which are unnecessary for my use.

I planned to finish it in a day. But got the design wrong: I thought the tree structure is always binary. In fact, one node could have multiple children. Also, the concept of forest can be implemented as a array of tree structure. Nothing special.

I worked from home this Wednesday, spend my work hours in Barnes and Noble nearby my home. Darn it the wireless network by AT&T is not free. But luckily my notebook detect a “unsecured network”, it’s slow, but free.

The 2nd day at work (Thursday) I realized I should have designed it to allow multiple children. But at least I got it right on Wed the visitor pattern shall be used. 

I demo it to Christof. He suggested me document the code. And think what should be the next step.

 

Sussie Sussie

Last night my last dream ( I probably had dreamed a lot at night. ) was all about Niuniu, my cat.  I called him, and searched for him everywhere. He is, as usual, hiding underneath the bed, or behind my bike. Poor thing. This morning right after I reached work, Keith came across my cubicle. I said hi, but he didn’t move on. Then I noticed he is holding a creature!!! A small tiny dog! Keith introduced for her, This is Sussie. She is all black with curly long hair all over the body. Her eyes is soft. She is so cute. But she refused to get into my arms. She seems very comfortable in Keith’s arm. Gorgous Sussie.

 

Orthology search, why are we interested in it?

To clarify some terms

Difference of Orthologs and Paralogs

Two genes are to be orthologous if they diverged after a speciation event.

Two genes are to be paralogous if they diverged after a duplication event.

(Source: http://genome.imim.es/~talioto/OrthologySearch/)

Two fundamental notions from evolutionary biology: orthology and paralogy, which describe the two fundamentally different types of homologous relationships between genes. Orthologs are homologous genes derived by vertical descent from a single ancestral gene in the last common ancestor of the compared species. Paralogs, in contrast, are homologous genes, which, at some stage of evolution of the respective gene family, have evolved by duplication of an ancestral gene. The notions of orthology and paralogy are intimately linked because, if a duplication (s) occurred after the speciation event that separated the compared species, orthology becomes a relationship between sets of paralogs (co-orthologs), rather than individual genes.

(Source: http://www.biomedcentral.com/1471-2105/4/41)

Homologous sequences are termed homologs and this term may be applied to both genes and proteins. Homologs look similar to each other and appear to share common ancestry but they may or may not display the same activity.

Analogs have similar functions/activities but heterologous origins.

Methods for determining orthology include the following:

• Best Reciprocal Hit (BRH)
• Conservation of Gene Order in corresponding syntenic blocks
• Phylogenomics (tree-based) methods

 

KOG/COG

We need a taxonomy structure to test the AIS API.

KOG: clusters of euKaryotic Orthologous Groups

COG: Clusters of Orthologous Groups

KOGs and COGs cluster eukaryotic and prokaryotic proteins respectively into groups containing sequences that are mutual best hits in sequence similarity searches between different species.

The COGs database includes proteins from prokaryotes and unicellular eukaryotes was substantially amended to include 66 species.

The eukaryotic orthologous groups (KOGs) include proteins from 7 eukaryotic genomes: three animals (the nematode Caenorhabditis elegans, the fruit fly Drosophila melanogaster and Homo sapiens), one plant, Arabidopsis thaliana, two fungi (Saccharomyces cerevisiae and Schizosaccharomyces pombe), and the intracellular microsporidian parasite Encephalitozoon cuniculi. In short, the KOG database includes proteins from H. sapiens, D. melanogaster, C. elegans, A. thaliana, S. cerevisiae, S. pombe, and E. cuniculi.

Source: http://www.ncbi.nlm.nih.gov/Web/Newsltr/FallWinter03/kogs.html

Year 2003

 

SIRS Event at HP Labs

This afternoon, we visited the HP Labs at Page Mill Road.

You can’t miss HP Campus along Page Mill road. (I always wonder if the “Page Mill Road” cames from the “Page” at Google. As a matter a fact, the whole Silicon Valley was initiated by HP, Hawlett and Packard, founders of HP, were two students at Standford University back to 1930s.

The first talk was given by the newly appointed HP Labs director Prith Banerjee. He was the dean of College of Engineering at U of Illinois at Chicago. This is the first time HP appointed the director from academic background. Prith did have two start-ups, though. His origin is Indian. He has a frank tone, and talks like from academic.

He gave a general overview of HP as well as HP Labs. HP has three major product sectors, one is personal systems, second is printing, third is data storage for business. Just to have an idea, how giant HP is, its Q2 FY 07 revenue is $25.5B. HP spends $3.6 B a year to its R&D. HP Labs has presence in Palo Alto, where HP headquarter is located, in Bristol, St. Petersburg, Tal Aviv, Beijing, Tokyo, and Bangalore. Maybe the new appointed director coming from Indian origin shows a sign the IT trend is moving toward Asia Pacific.

Next to Prith, Bernardo Huberman, a senior fellow, the highest rank in HP kingdom, gave a talk for a series projects undergoing in HP Labs, under the theme of “Harvesting Organizational Knowledge”. It was a fantastic talk. The ideas are interesting, the presenter passed the messages across to us well. I can notice people are paying much attention to his slides, and they move around their shoulder so as to see the charts on the screen. He has presented:

1. How to find an expert in certain domain within an internal corporate network. The motivation is, at HP there are so many consultants, researchers, engineers. If one got a question, how does he quickly find out who is the right person to ask for help.

They tried to find the expertise by analyzing the access patterns to information such as documents, blogs, web pages, of the HP consultants.

Also, they search inside the PowerPoint slides so as to imply the author of the slides are experts or not.

Their research prototype is called PeopleFinder2.

2. How good is Wikipedia? Wikipedia has been growing faster and faster. How good is the quality of a page content?

They have done some experiments showing that the more time a page is edited, the better the quality.

3. How to predict the future? It has much market value, as the decision maker in the business wants to know what will happen next.

4. How to draw attention of a group, in condition that there are so many distractions on a web page. He gave a metaphor, to draw attention to a stop sign in a dessert is no problem at all. However, to draw the same attention to a stop sign in Tokyo business district at night is almost impossible. As it is badly light polluted.

Project page: http://www.hpl/hp.com/research/idl

After this presentation, we were divided into groups to observe some project demos.

1. Data center cooling system. How to keep the temperature down in the data center. Imagine 50 people sit on top of one chair!

2. Memory spot and Near Field chip. The photo now is able to embed an audio at the size of 3 square mm chip. The photo “sings” when the reader scans in the chip. The data transfer rate is about 32k bps. Not bad. And it can holds up to 20M data.

A book page is now embedded with a 25-cent coin chip. The associated web page is displayed on the reader screen when it scans this chip. However, this chip can holds up several byes.

3 Simulate a high end Hollywood movie quality projector by coordinating 4-6 household projectors.

4. A boring project about supplier/retailers behavioral study.

5. Quantum Study Lab. I feel they are quite a good team. It is confirmed by Bill.

Our last stop is at Hewlett /Packard’s office. The office was kept as the same as it was back to the time when he was working at HP. On the wall, there was even a photo taken him with Queen Elizabeth II! On his desk, are many different kind of currency notes, coins. I found Singapore Dollars! haha. Tourist are all the same all over the world. :-p

The reception was pretty good. There was a trivia quiz. The winner team was given the right to visit and play in their gaming lab.

 

Another day of Paradise

Got an email from Christof after I came to office. He asked me, in my last email, “what do you mean by ‘doubts’ ;^) ” At first I thought I had a typo, as I am always not confident enough for my spellings. I rely so much on the spelling tool. Outlook doesn’t mark the wrong spellings, unlike Gmail, or WordPress! hee hee, Christof is marked by WordPress as a typo, though. I checked the online dictionary, doubt, unclearness about a fact, etc. And it can be used in plural form. So, not a typo. I guess he is too bored in Berlin on the conference.

Craig is a software visualization guy. I asked him for pointers to software visualization tools. I need such a tool that it can visualize the Java code into inter-related UML class diagrams. I clarify with Craig I don’t need a real time stack trace tool; I don’t mind 2d or 3d display. We searched for some promising links. I am giving ArgoUML and VisualizeIt a shot. See whether they can scale to large size software or not.

None of the tools works! I tried ArgoUML, tried SHRIMP, tried Visualize It! ArgoUML doesn’t have a feature to draw class diagrams given java files.  Visualie It took me long to download and install. Finally it asks for license file! SHRIMP, the java web start shows a simple demo, the Eclipse plugin doesn’t show up in Eclipse’s available perspectives. I stop searching this kind of tool, after writing SHRIMP support an email.

 

Understanding AIS API

Today I started diving into the AIS API code to understand how the active information store concepts are realized in it. I first read the design notes superficially, then realized I can absorb nothing from it. Reading the code is much more grounded. I first analyze the SchemaGenerator test program. From that test program, I followed the threads of relevant class declarations. Know what’s InforSource, WorkSpace, and AssociationType. I got many questions, got to ask myself tomorrow and probably ask Horst again.

Horst sent me two updates of the whole API Eclipse project. I have to unzip them and update my local copy of the API. I suggest using a version control tool for this multi-developer project.

 

Tele-meeting with Horst

Today Horst and I had a long meeting over the phone for our project. He is such a hardworking fellow. After 3.5 hour meeting, he still planed to work for another 1.5 hour until 2am!

He walked me through the test programs, gave me an overview of the most important concepts and classes in the API. This is the first time I work on API level of Java Programming. I am quite eager to it. He told me it took him and Christof 2.5 months to finish the first draft of API. WOW. How soon would I be able to finish it? Horst is very patient.

At first I can’t even compile the whole Eclipse project. It’s the path setting problem, as usual. After the paths are fixed, the program still won’t run. This time the connection to the DB2 server can’t be established. We thought over the possibilities for the problem. First, DB2 database manager is not started. Second, no relevant database is created. The setup guide I followed didn’t mention creating a database instance. Horst modified the set up guide. Still I can’t connect to the database. The error message says authentication failed. We then looked into the configuration file where the credentials for connecting to the database server is specified. I changed the userid/password from dbadmin to db2admin, then it works! The Eclipse project runs and return the expected results.

Horst then told me what the test programs are about. It was long explanation, as he need to refer to the API itself to explain to me.