Vagaries
Aug. 25th, 2007
05:40 pm - another watch out
After Network World's top 10 listing, Byte and Switch puts 3Leaf in it top 10 storage virtualization companies to watch for!
very cool.
In other news, I am thinking about going to Burning Man next week. Still need a few more co-travelers to reach critical mass.
Aug. 17th, 2007
01:03 am - yacss! (yet another customer support story)
Owing to the recent mail-theft fiasco, I ended up receiving some items twice. So I wrote to customer support of one of the big online
stores asking of a way to return back the extra item that I have NOT paid for. The response I get is that they are terribly sorry but they can't
take it back as I didn't make the return within 14 days of purchase.
I am not quite sure what to do next...
Mar. 29th, 2007
08:33 pm - mailing woes
I lost a few mails on my gmail a/c this week as exceeded my storage quota ( I am a popular guy). So, now I have to
spend some time deleting mails from the a/c that didn't even had a delete button a while back.
And this happens on the heels of Yahoo announcing its unlimited storage mailbox!
Mar. 24th, 2006
12:27 am - pedometers
I recently restarted the morning jogging routine with my roomie. I was looking at some pedometers to get an idea of how much distance/calories we were putting in. After looking up a few cheap and slightly expensive pedometers, I found a satisfactory software solution with gmap-pedometer. It uses the google map api and adds some simple functionalities.
Here's the route we started with - a block around our apt . I really like this tool. There isn't any useless info, just the basic things I want. And its great to plan new routes.
For the tech-oriented, check out the Acknowledgement link for some reference material about the site design.
Jan. 24th, 2006
11:32 am - SimNow
AMD SimNow is muchos fun, especially if you have the full licensed version.
Amazing level of control. Do try sometime.
(
And all those who have not caught up with my life recently:
. all is well in Santa Clara.
. missing friends, meeting some old ones
. missing co-workers, meeting some new ones
. cold is gone, cough is still around
)
Jan. 11th, 2006
08:35 pm - Rescue operation completed
Among the screwees were such diverse items as pcmcia, /dev/pts, and apic.
Moved things back to bochs.
That was fun, should try it again someday
11:09 am - you live, you learn
Depression caught me up again last night and I thought of beating it out by getting some
work done.
With half a sleepy eye on specs, reprogrammed some hardware, rebooted - and roasted up my whole setup.
Rescue work in progress. No casualities reported yet.
Dec. 3rd, 2005
11:46 am - Yahoo Mail Authentication ...
Yahoo has started asking for captcha confirmation before sending mails.
You have to type in letters from an image before you can send a mail.
It is supposed to help block spammers.
Has anyone else noticed it? I find it very irritating.
Oct. 21st, 2005
10:58 am - Education
Old Man Pratchett said once that Education is a bit like getting an embarrasing sexually transmitted disease. It makes you unsuitable for a lot of jobs and you get this strange urge to pass it on.
I did my part of 'the passing on' these past few weeks. The topics were UNIX Internals - System Architecture and Memory Management (yes, again).The audience was good and the material can be found here, here and here. [warning: Evil ppt format. pdfs are not uploaded yet]
For the record, the count of my lectures on linux/hpux/unix has now moved on to double digits.
Jul. 18th, 2005
06:50 pm - Lateral Programming ...
Spent some time today playing with Duff's Device and then followed it up with some other unusual embedded systems and compiler performance tricks.
Digging through some more artciles at DDJ now.
cool stuff !
Jun. 9th, 2005
10:50 am - Yahoo! Photomail
Whoops!
We're sorry, but your computer doesn't currently meet the system requirements for PhotoMail. To use PhotoMail, you need to be on a Windows 2000 or Windows XP PC running Microsoft Internet Explorer version 5.5 or higher. Click here to download the latest version of Internet Explorer.
In the future, we hope to offer PhotoMail for a wider range of operating systems and platforms. In the meantime, you can always share photos using the "Attach Files" button in Yahoo! Mail, or check out Yahoo! Photos for easy ways to share your digital pictures with friends. Go to Yahoo! Mail
May. 10th, 2005
07:07 pm - code browsing ...
What are the chances that you read 'ass_hole' in the code and
think number of holes in the address space?
Apr. 20th, 2005
04:27 pm - USENIX 05
Some *very* interesting talks/papers here ...
http://www.usenix.org/events/usenix05/te
Feb. 22nd, 2005
02:07 pm - Groupware Bad
Culled from JoelOnSoftware ...
Jamie Zawinski On Groupware
"Groupware" is all about things like "workflow", which means, "the chairman of the committee has emailed me this checklist, and I'm done with item 3, so I want to check off item 3, so this document must be sent back to my supervisor to approve the fact that item 3 is changing from `unchecked' to `checked', and once he does that, it can be directed back to committee for review."
Nobody cares about that shit. Nobody you'd want to talk to, anyway.
-.-
And on Free Groupware ...
With a product like that, there was going to be no teenager in his basement hacking on it just because it was cool, or because it doing so made his life easier. Maybe IBM would throw some bucks at a developer or two to help out with it, because it might be cheaper to pay someone to write software than to just buy it off the shelf. But with a groupware product, nobody would ever work on it unless they were getting paid to, because it's just fundamentally not interesting to individuals.
So I said, narrow the focus.
Your "use case" should be, there's a 22 year old college student living in the dorms. How will this software get him laid?
-.-
Feb. 19th, 2005
11:47 am - Quoting Jim Gray
From an interview with Jim Gray:
Q: I understand that your PhD thesis looked at parsing theory. What led you to switch from parsing to databases,
and what led you away from the theoretical side of things?
JG: Well, I m sort of interested in everything, and I felt I could have written a practical thesis or a theoretical thesis,
and I was, I guess, kind of in a hurry to get out. I did my PhD in a year and a half. If you re in hurry to get out,
you do a theoretical thesis.
The reason for that is that if you do a theoretical thesis, then you do the theorems, you do the proofs, and you
describe what you ve done, and you're done. And in a systems thesis or practical thesis, you have to build a
system and then you have to explain what you did. That kind of thesis is twice as much work as a theoretical
thesis, because you have to build a working system and then you have to write the thesis about it.
Dec. 23rd, 2004
10:43 am - Computer Architecture - HiPC keynote

Attended Prof Arvind's keynote address on "Rethinking Computer Architecture
Research" at the International Conference on High Performance Computing' held
in Bangalore (Dec19-22). The main focus of his talk was how to efficiently validate
and realise any proposed architecture changes like cache coherency algorithms
or reorder buffer implementations.
In reasoning out why computer architecture research is in a bad shape and
how most of the researchers are focussing just on incremental research like
a performance gain of 2-4%, he took a shot at the computer architecture bible -
Hennessy-Patterson's Computer Architecture: A Quantitative Approach which,
though being very successful at teaching computer architecture, has set the
trend of Quantative analysis in research too. Most of the research papers these
days concentrate just on quantitative analysis and do not even mention
implementation details or even feasibility. He then went out to describe his RTL implementation tool - bluespec
and how it simplifies preliminary verification and implementation modelling.
A few months back, I started studying caching / coherency techniques and faced similar problems.
Some of the algorithms were beasts for implementation and lot of papers never ever mentioned how their particular
scheme can be implemented. And you will be surprised by the number of papers published just on cache coherency.
Had a very frustrating time there :(
I am still looking for some good material on hardware implementation of caches (Got Any?)
Dec. 9th, 2004
03:52 pm - Follow-up on Nuclear AA Size batteries ...
More disappointing news from techdirt, people are accepting the state of batteries to remain the same and
working around the problem.
~.~
From techdirt ... Forget Advancements In Battery Life -- Just Decrease Power Consumption
Contributed by Mike on Thursday, December 9th, 2004 @ 01:14AM
from the ah,-so-simple... dept.
The issue of battery life is a big one in the consumer electronics world -- especially for mobile phones. While a lot of research is being done on ways to increase battery life, there is no Moore's Law for batteries, and real advancements are few and far between. Basically, the more power you pack into a small battery, the more like a bomb it is. It may be possible to pack a super long life battery, but it involves a high risk of exploding -- which is something most people prefer to avoid. So, if you can't extend the power output of a battery, why not work on the next best thing: lowering the power consumption of the device? That's exactly what some researchers in Canada appear to be doing, as they claim they've produced an analog chip that requires only 1/100th the power consumption of existing digital chips. There are a lot of questions this raises -- and I wouldn't expect to see (as the article suggests) phones that only need to be charged once a year any time soon. However, it does point out that there are other ways to approach questions like battery life on devices.
~.~
Anyone for a nobel ?
Dec. 4th, 2004
01:43 pm - You Goofed up, I am so Sorry!
Says New Scientist, Computer glitches would be a lot less annoying if the machines were programmed to acknowledge errors gracefully when something goes wrong, instead of merely flashing up a brusque "you goofed" message.
In the article titled "Polite computers win users' hearts and minds", Apologising computers are predicted to be much better in winning new users than the good ol' "Core dumped, your fault" types.
I tried re-writing a few error messages myself and do sound a lot better - and much more fun.
$ ls /non/existent/file
Oh User, forgive me for my ever erring way
Help me find the path again, I never meant to stray
$ rm /very/important/file
Forgive me O thrower of commands for doubting you,
For thou, User, art good, and ready to forgive;
but do thee REALLY want to screw thyself?
The article goes on and discusses incomprehensible messages like "HTTP 404 - File not found". And I tend to agree on this point too. Its much more fun with the apologising computers. These guys surely have seen Marvin doing 404.
Oh, And I like this article for another reason too. It mentions a name. Eric Horvitz. Manager of Microsoft's Adaptive Systems and Interaction group. Responsible for creating MSOffice's irritating, stupid, ^&^%#* "paper clip" assistant. Another addition to the "Kill Bill" list :)
The article ends with a profound statement by the researchers - "You don't have to understand how people really feel. Software designers just have to develop the attitude that the user is always right."
My Boss has competition.
Apr. 21st, 2004
10:51 am - C++
I have been working in kernel for quite sometime now and therefore been programming mainly in C.
Looks like I am getting a bit rusty with C++.
So, looking for suggestions for a C++ project. The aim here is to use most of C++ features and constructs.
Any suggestions?
Apr. 15th, 2004
11:00 am - Crossover Office ...
Found this cool app ... installing m$office on your linux box and even integrate it to KDE/GNOME interface.
Using the demo version now ... way cool.
Check it out here
Navigate: (Previous 20 Entries)
