Aging Laptop

(Just to remind anyone who wants news of Hui Ning, my new daughter — head on to my other blog. And if you decide to venture there do be aware that there might be TMI at times...)

My Fujitsu laptop's showing symptoms of age, even though I've only had it for about a bit over two years. Then again I showed it no mercy, using it for >16 hours per day frequently, coupled with the temperature in M'sia...

Bottomline is, the fan refused to work, even at startup, about 2 months back. I get a loud beep at boot-up and an erroring declaring that the fan system is off. I was trying to knock off some projects before the baby arrived, so I continued using the laptop anyway. It probably wasn't a good idea. A few days back I found that after about 30 minutes of usage (without air-conditioning), the graphics card seems to go all wonky: in Windows, the screen just froze. I restarted the machine, and I see vertical stripes interspersed with the usual BIOS startup splash. In Debian though, X.org seemed to work OK, but come shutting down time, the console messages are a bunch of garbled text.

Uh-oh. Sounds like some stuff are getting fried. Back-up time. And only use the machine with air-conditioning on when it's absolutely necessary (like uploading Hui Ning's photos and videos, heheh.)

Then today, immediately on startup, the vertical stripes were already there in the BIOS startup. Windows is rendered unusable as even the login screen can't be displayed. Debian though... ok so the startup console messages look like some pig-latin encrypted code, but in due time X.org starts up fine and I'm here writing this blog post while backing up whatever else I should be backing up. (Why do I feel like writing "We cannot get out. They are coming. We cannot get out."??)

This laptop goes to the service centre tomorrow — hopefully we're not too late.

Welcome, Baby Hui Ning!

Baby Hui Ning 慧宁 arrived roughly a week earlier, at 12:58pm on 30 July 2008. My water broke around 6a.m., so I guess my labour wasn't too long for a first child...



I'm still pretty exhausted (and time-constrained!) to blog much -- you know the drill ;-D In the meantime, here are some photos.

38 weeks

Until I saw myself on film I had no clear idea how big I've actually become....

1.5 weeks to go (assuming we're following strict due dates) :D !

Double Yolk

I was making a quick snack and out popped this twin of a egg into the saucepan. (Sure, call me kampung to even want to blog about this.)

iMac in the house

After the desktop PC conked out yet again (the power supply this time), we've had it with PCs and Windows.

Result: We went out to Sunway carnival and got a spanking new iMac.



We love it. Lotsa things to explore.

Now we've got three OS'es in the house. BTW the wallpaper there is Semicolon from Vladstudio.com. (Also, the panda didn't come with the Mac...)

Here in My Home

Nice song. Fun video (and I don't just mean identifying the personalities without looking at the credits). Soppy and too-good-to-be-true dream, though. Yes, I know I sound cynical. But I can't help the warm fuzzy feeling (and the sour cynical feeling) after listening/watching this.

The MP3 and QuickTime video can be downloaded for free at MalaysianArtistesForUnity. So go on, watch the video, download the mp3, embed the banner/video in your website/blog, spread the word. If only because it's a nice song.



Update: I badly needed a LaTeX fix, so I re-typeset the guitar chords (original .DOC version available at the site above). The LaTeX-ed guitar chord + lyrics sheet can be downloaded here.

Update 2: This is weird. My baby kicks whenever I play the song. HARD. Yes sweetie, mommy hopes that this'll be a good home for you too. A better one than now anyway.

Attack of the Sudden Visiting Researchers

(This is the last in a series of ranty posts, about the events of a particularly... "exciting" week. Yeah, I hear many sighs of reliefs — or is it just one since I really only have a readership of one? )

'Twas the morning before the ScienceFund project presentation, and I was waiting for CC to come to the lab so that I could set up the demo system on his laptop.

When suddenly...

Mr T appeared and told me to quickly come to the office, 'cos the main office just made a surprise call saying that a team of visiting researchers from a Thai university are coming down to each of the labs, so, yeah, get ready to receive them and show them around.

"When's that happening?"

"In, uh, like, 5 minutes' time."

Well thank you Main Office for the short notice. Though I guess everyone forgot everything else with the ScienceFund projects for the last one and a half weeks, so it's probably unfair to blame them.

Receive the visitors, eh. (And then what?) Mr Tan the word was to brief them about the research we do. Sure, get someone to talk, but we were in a bit of a fix: Dr T wasn't around. Dr R wasn't around. I was the only RO around in the lab at the ungodly hour of 11am (hey everyone was exhausted the whole week scrambling for the presentations and reports). Mr T had already dragged out Dr S, but she's into linguistics and not really crazy about the "computational" part — and on seeing me she said "Oh good, you're here. You can do the talking then."

Ack.

Then I remembered about the brochures for last year's open day, and luckily we still have a few copies lying around in the office. And then it's scurrying back to the lab to hopefully find a big enough space for 7–8 people to huddle around a computer monitor (Dr S' suggestion) and to hopefully locate some online demos that we can show (Dr C from next door's suggestion).

They came.

I gabbed. Tried to gab. Then showed them the online machine translation prototype, the multilingual lexicontology (yeah it's a mouthful, but the real project name is even worse) and the (not quite reliable) sense tagger.

Surprisingly the visitors were deeply interested (and even impressed with the sense tagger)! The questions they asked showed that they were no strangers to NLP. They even left name cards (which I yet again embarrassingly told them I didn't have one since I'm a mere down-trodden RO. ) En A who escorted them around said that of all the labs visited thus far, this lab was the longest time they spent.

On a side note — En A (sorta cluelessly) remarked that this was also the most run-down/unimpressive (something to that effect, dunno the exact word used 'cos this was re-told to me later by Mr T) lab they've seen that day. To which Dr S retorted "Well this is the lab where the people do real work!"

Microsoft Office 2007 to support ODF?

If what's reported in this PC World article is true, I'd be so happy. I can then happily work away on OpenOffice and e-mail them to all the other Microsoft Office users without converting them and getting a bunch of tables and bulleted lists wrangled in a the process!

Attack of the LaTeX Typesetting Job Windfall

Well since it was a windfall I probably shouldn't call it an "attack" – but as we raced towards the end and with other stuff happening, it felt like it.

It was nearing mid-afternoon on Sunday and I was in the middle of downloading the latest versions of Apache Tomcat and PostgreSQL so that I could install my project system on Windoze, when I received an SMS from a USM PhD candidate. His conference paper was selected to be published in a Springer journal, but the catch was that it must be in LaTeX. He tried to learn it that morning (I greatly admire his courage), but gave up by afternoon. As it was a tight deadline he decided to offer me a bounty to do the job.

I'm always up for a LaTeX fix (it's the high I'm after not the $$, get that straight! ) so I said Yes. When's the deadline though? I was hoping that I could finish dealing with the ScienceFund project presentations first.

Tuesday, came the reply.

Yikes.

I decided that I still needed my LaTeX fix. Besides, this was my first paid LaTeX assignment! (Did I mention that it's the high I'm after and not the $$?)

The LaTeXing itself wasn't too difficult, although the conference organiser's failure to post the correct version of style files to be used made it a bit frustrating. We were told to go hunt the correct version down at the Springer website instead. Also a couple of not-so-trivial table formatting, but luckily the trusty LaTeX Companion book was ever helpful. In the end what proved to be my downfall were typos. With the deadline fast approaching we went absolutely frantic with the to-and-fro e-mailing and SMS-ing. These keep turning up till 11.59pm on Monday night, when I just couldn't hold my eyes open nor my head up anymore, and had to call it a night. Luckily my client didn't come across any more typos after that, and he submitted the paper in the morning.

(I'm really sorry about all those typos, J!)

In hindsight I probably charged a bit too low for so much effort, especially with my tight deadline and packed schedule at the time. I guess I should consider it as a First Client Discount though.

So if you have any LaTeX typesetting job (for a fee), why not try dropping me a line? It'd help a lot if you also attach existing text in soft copy (Word etc) if any, as well as information about the formatting or style guidelines required. A link to the conference organiser's or journal publisher's web page providing such info should do. A comfortable timeline for both you and me would be one week at least, so that we'd have time to review the outputs. Anything shorter than that and I'd either have to turn down the job or charge extra.

Please note that I will not typeset your thesis for you, though. I'd be happy to offer advice and answer questions on how to use the USM thesis template, or even design your university's template for you (price negotiable), but not typesetting your thesis itself.

And remember – I do things like this for the fun of LaTeXing, not for the money! Me, crazy? Sure, why'd you think I name this blog what it is now?

Well cheers, folks!

Attack of the Sudden Nation-wide ScienceFund Project Evaluations

This, dear reader, is going to be the first in a continuous series of ranty posts. It's been an... exciting week for me.

Here's the chronology:

Apr 29, 11am:
The first inkling I had, was a fleeting comment from C.C. about his having to go down to Putrajaya on 7 May to present a ScienceFund project he's involved in. I thought his project was picked on by random, because that's what happened to another researcher before. Then he told me it's actually because MOSTI is requesting all the 1st cycle projects (ending May '08) to this evaluation/presentation workshop, so that they can decide whether they'll release the last payment.

O_O

What about my project on the multilingual dictionary then? Haven't heard anything. But I suppose all correspondences would've gone to Dr C's mailbox and she's on extended medical leave...

C.C. checked against the list of projects he had and told me my project wasn't listed. So it doesn't concern me then...

Apr 29, 1.15pm:
Back from lunch, and found a note on my table from Mr T. Well well weeeeell, whaddaya know. There's a second list of projects circulated just this morning and my project's slotted for 7 May, too. Talk about short notice.

It's all a bit spooky though, since I just finished all the programming and put the servlets online last week, and was actually going to update Dr T and Mr T via e-mail anyway.

After a quick discussion with Mr T, he went into doc compilation overdrive mode (harrassing Bendahari and RCMO for the respective docs, photocopying whatever docs and receipts he has on hand etc), and I went into PowerPoint overdrive mode. The best part is that many academics are on leave one day or another this week, since Labour Day is this Thursday, and – wait for it – almost all USM clerical staff are attending a seminar this week!

At least we didn't have to rush up on the technical development stuff. Also, Dr T will be around Cyberjaya/Putrajaya on 7 May (he'll be at a seminar in Port Dickson until 6 May) to present the project, so I won't have to travel down. Although Sara commented that if I went down nobody'd dare ask too many questions on seeing my bump...

Apr 29, 6pm:
MOSTI just won't disappoint – Dr T e-mailed to say that there's been a THIRD list and now one of his projects (in 2nd cycle) is also slotted for presentation, also on 7 May!

Apr 30:
Got most of the slides in order by last night, so today it was mainly checking over them and help brief H.H. and C.K. for Dr T's project.

Holiday tomorrow – don't care, am going to kick back and relax.

May 2:
It wasn't until late evening when the financial details came in, so I incorporated those at night. Also got together a bunch of documents to be printed out in hard copy first thing on Monday. Apparently there was also some RSVP slip (e-mailed out today) that had to be returned to MOSTI by... today. Eugh. Hopefully Dr T saw it; anyway I SMS'ed him to check his inbox in case he didn't.

May 3, around noon:
Dr T called in the morning – apparently he's already in Port Dickson since yesterday! Quickly called Mr T who's back at the office to continue all the photocopying. We'd have to send the RSVP on Dr C and Dr T's behalf; and hopefully it'd land in the right inbox since apparently we need to RSVP to different people for different panels...

Suddenly remembered that I'd better check with C.C. that he's sticking around in Putrajaya long enough to be in time to pass the projects to Dr T! More SMS frenzy. Turns out that they have the first slot 9.00–9.30am and plans to fly back to Penang by noon! More SMS frenzy to Dr T to see if he can make it to Putrajaya in the early morning.

Then decided to forget all about it as Dr T couldn't respond just yet (he's at the seminar remember?). Went to RedBox to semi-celebrate husband's birthday, semi-release stress (both of us needed a break), before heading to the World Music Festival (will blog that later! ) in the evening...

May 3, 7pm:
Was at the Botanical Gardens for the World Music Festival when C.C. sent an SMS saying that Dr T has arranged for him and H.H. to stay overnight at Dr T's place on 7 May – so no problem for him to hand over the docs to Dr T then!

Ah bliss.

But little was I to know...

May 4, 11am:
Got a call from Dr T asking me to install my system on C.C. or H.H.'s laptop, just in case there's no internet connection at the workshop venue. And we all know how very true that can be.

Yikes. I've never tried installing it, especially PostgreSQL with all that huge amount of data, on Windoze before!

May 5:
Stupid. Windoze. Wouldn't. Print. To the lab printer! Eventually I had to delete the printer, restart Windoze, add the printer again, for it to work. Bah. Well by noon I finally managed to print whatever docs I needed. Finalised the presentation slides and docs with Mr T in the afternoon, and he carried off the docs for binding/compilation.

I had to switch tracks a while and get rushing on a LaTeX typesetting job due the next morning, or today midnight... (more on that in another post.) Then spent the whole evening trying to work out the system installation/config/setup details on Windoze, till midnight. I'm getting worried. Them conceptual vectors data are really huge and seem to take a looooong while to load into the database.

May 6:
Sudden Attack of Visiting Researchers (more about that in a later post) in the mid-morning rather threw all of us off-balance. It was after lunch when I could start trying to set up the system on C.C.'s laptop. To my surprise it took less time than I expected – a bit over 2 hours. However PostgreSQL just managed to start up once: it refused to start again after I shut it down, so that I could show C.C. how to launch everything. As we were all damned tired I told C.C. and Dr T to use the demo video I created with CamStudio. The sense tagger can't run locally off the laptop anyway if the internet's inaccessible since it relies on our FDG server.

I hit home after burning the presentation slides and video onto a CD and passed everything to H.H. and C.C. I'm beat.

May 7:
So this is the day – but I slept in late anyway. Too, too tired. But I still made it to the lab relatively early (10am) to make sure that the server was still running. In the meantime Dr T called to confirm a few points in the slides; then C.C. called to tell me that while they do have Internet access there, the firewall blocks all requests to ports other than 80 (HTTP), both ingoing and outgoing.

Huh?! How stupid can that be?? They've probably just voided out all Apache Tomcat-based applications, and then some.

C.C. also appeared to have found what's wrong with the PostgreSQL installation; apparently I made a typo in postgres.conf! Once that's corrected, everything's running as expected – well except for the sense tagger, since it sends requests to port 8080 on the FDG server – and that's blocked. Meh. Luckily we still have the video.

In the afternoon C.C. SMS-ed back to say that they've finished all the presentations and the panels seemed satisfied. What a relief! (Although for some reason, the MOSTI people requested to bring back all the docs for my project with them.)

And now to bed! And now to bed!!!

iGoogle Artist Themes

You can now customize your iGoogle page with themes by contemporary artists. They're all interesting, but some are definitely acquired tastes. :-D

A website of colours

I'm really lousy when it comes to picking colours, especially if it's fiddling with RGB code to produce anything that looks remotely interesting. Luckily for me though there's COLOURlovers. Community users (called "lovers") post their concocted colour palettes, and can be downloaded as GIMP, Photoshop or HTML etc format. The blog posts are interesting reads, too. It's all really very nice.

More kitty fluff

Awwwwwwwww me want!!! Me want!!!

Fractal Google Gadget

Mmmmmm. More eye candy...

Wow. Hibernate worked!

On my Fujitsu laptop. Running on Debian. Wooow.

Debian yammered about starting the snapshot process (checkpointing, Hussein sez) and proceeded to write n number of pages. Once that's done, the laptop powered down, and I went out to dinner. When I powered on the computer again, what greeted me was the GRUB menu. And I was thiking "well so much for false hopes..."

Once I booted into Debian though, it started to yammer about resuming from n pages of image (checkpoint-restore, Hussein sez) and voila! It's back to exactly where I left off!

Both the checkpointing and restoring took longer than in Windows, though.


ps: Apparently Hibernate is not expected to work usually:

<obiwan> hey guys
<obiwan> 'Hibernate' actually worked for my colleague.
<ahu> .. once
<ijuz> probably some new drug
<obiwan> No, it really worked

Hahahahahahaaaahahaha! :D

Viruses never looked so good

MIT grad student Alex Dragulescu was asked to turn viruses, spams and spywares into graphics with visualisation algorithms -- and the eye candies that emerged are simply c00l. (The thing at the top is MyDoom.)

The other blog -- on Fluffy Baby Stuff

My friend Hussein was expressing his trepidation of me spouting fluffy baby rants on this blog once it comes into this world come August. (I suspect he's also rather apprehensive of having cases of TMI.)

Fear not, friend. I promise that all baby fluff stuff will be posted at this blog -- since such fluff doesn't really "fit" here anyway (think baby talk, brand comparisons of baby cribs etc). So you don't have to read all about it if you don't want to, simply don't subscribe to the other blog feed. ;-)

One thing, though: kitty fluff do belong here! Any questions?

Lost Li'l Bird

This little guy got lost and made his way into our house the other night.

It was getting quite a bit confused, chirping, hopping and flying around the house in frustration for not being able to find its way out of the house again. Eventually we had to turn off all the house lights except the porch light to guide it out again.

HAL's Legacy

Quite a while back, while I was surfing for "fun" bits of information to be included in some course notes, I came across this e-book version of Hal's Legacy.


Hal (or more formally, Hal-9000), as you might recall, is the supercomputer in the science fiction book/movie classic 2001: A Space Odyssey who (yes, I insist, who) eventually committed murder in an endeavor of self-preservation. Hal's Legacy is a great book for students interested in artificial intelligence, natural language processing, robotics, or just geeks in general. And it's online in its entirety for free!

Enjoy!

牛奶＋柠檬＋蜜糖＋薄荷牙膏

肚子里的小瓜总在我进食后呱呱叫，胃又好胀好胀气，心烧了又烧……

四处搜索，多番尝试下，竟然发现以下几个方法可行：

1. 蜜糖泡水，再加一片柠檬。（冷热随喜）
2. 牛奶！反正牛奶是最容易汲取钙质等营养的饮品。
3. 饭后勤刷牙，尤其用薄荷牙膏——据说口腔清爽可以减少胃酸分泌。