Nekochan Net

It's a bit of a pity, since IRIX once was the most popular UNIX platform in the biology market. Back in olden times (late 90s through around 2002 or so), SGI themselves actively ported and optimized several bioinformatics and molecular biology tools for IRIX. For applications that took advantage of multi-processor machines, like BLAST, FASTA, or CLUSTAL, the SGI optimizations were pretty impressive. The lack of a current Java and a modern, well-performing browser hurts quite a bit. Most of the scientific tasks that you mentioned are well within the processing power of an O2 (unless you're doing whole genome comparisons), but without a good Java implementation or a browser, IRIX does get left behind. In a current scientific environment, an O2 still works nicely as a front-end to things running on larger systems or on clusters, but mostly as a glorified X Terminal.

Edit : Went looking around - what a bunch of absolute crap software !! And it doesn't seem to matter if it's Irix or not, just junk ! I feel sorry for you, actually. If I were in this field, I'd be spending weekends at the range, practicing with the AK-47. Looks like this area is ripe for some decent projects !

A lot of the collaboration setups nowadays are web based. Sure, you can use e-mail, but sometimes other options work better.

Heck, Nekochan is web-based.

_________________
Damn the torpedoes, full speed ahead!

Systems available for remote access on request.

(single-CM)

And nekochan works beautifully on Fireflop 2

The stuff that doesn't work, is crap. So don't go there.

In fact, being a glorius terminal is one of its main roles, I need it to talk with some big clusters and it also manages the automatic backups, etc. My other machines are Windows based and the O2 is a very nice Unix box if you do not want to have an Apple, like me It is also great for the MP3, mails, etc., and I find extremely satisfying to work with the girl .

About the crappy software, it is just that, although it works after all. You can buy a big commercial suite that does almost everything, but I prefer to work with free software because I do not want to be the slave of a particular company. Moreover, the price of those suites is usually crazy. The labs can pay them, but then you move to another place, they use another suite and all your files are useless. Luckily, I also have great free programs, like VMD.

http://www.ks.uiuc.edu/Research/vmd/

.

_________________
_Betty Blue_
R12000A 400 Mhz; 1 Gb RAM; 72 Gb 15K HDD; IRIX 6.5.29
CrystalEyes; Dial Box; O2Cam "ZEYE"; external Toshiba SD-M1711 DVD-ROM; Octane speakers;
Lock bar; SGI microphone.
Mods: PSU Noctua fan; internal Toshiba SD-M1401 DVD-ROM; Adaptec AIC-7880P SCSI card.

_REKIEM_I7_
Seasonic X 1250W PSU / Intel I7 2600k 4 x 5,00 Ghz / 2 x Gainward 2Gb GTX 560Ti Phantom 2 / 32 Gb DDR3 / Intel x25-M 160 Gb SSD and 2 x 1,5 Tb HDDs
_Lazarus_
2 x Intel Xeon MP Gallatin 3,00 Ghz with 4 MB cache / Zotac 512Mb GT430 / 12 Gb DDR266 ECC / 4 x Maxtor Atlas 146GB 10K V U320

I don't envy you. Of the programs you cited, three out of four use Java as some sort of interface ? That's just sick. And I thought GTK2 was bad ... The clustal2 command-line version makes object files but fails to link. You might get that running ...

Okay, I'll backtrack. If you work in any normal field you'll be able to find applications that run in Irix. Regular desktop, graphics, design, mechanical, circuit design, most everything. Are you sure there aren't any better programs than that shit to use ?

...and that's another pity. (JH now slugs some whiskey and switches to his "old timer" voice.) Back when I was working with many of the same analysis tools, people used things like Tcl/Tk to write wrappers and cross-platform GUIs around the basic scientific algorithms, and that approach worked very well, with lots of GUI goodness and excellent performance. From a user's perspective, Java didn't add a whole lot to the experience, aside from the "Gee, look, we can run this app (slowly) within a browser window, too!"

It's kind of lost in the dim ages of history, but on the enterprise side, Oracle's GUI installer and management tools were implemented in Tcl/Tk on UNIX platforms until Oracle 7 or so, and they generally worked quite well. When Oracle switched to Java with Oracle 8, the difference in quality was very noticeable. At the time, it felt like a hard kick towards Windows or Solaris and away from other versions of UNIX.

Sadly, I have to live with those applications, there are no better ones unless you chose a suite from a company . I guess that many of them are made in Java because it is a quite popular programming language in the field, as well as Phyton. Most biologists learn one of these two. I code eveything in C or C++, but I am almost the only one. However, my skills are not enough to create my own software, hehe. Moreover, it would take a big amount of time and I do not have it :/.

_________________
_Betty Blue_
R12000A 400 Mhz; 1 Gb RAM; 72 Gb 15K HDD; IRIX 6.5.29
CrystalEyes; Dial Box; O2Cam "ZEYE"; external Toshiba SD-M1711 DVD-ROM; Octane speakers;
Lock bar; SGI microphone.
Mods: PSU Noctua fan; internal Toshiba SD-M1401 DVD-ROM; Adaptec AIC-7880P SCSI card.

_REKIEM_I7_
Seasonic X 1250W PSU / Intel I7 2600k 4 x 5,00 Ghz / 2 x Gainward 2Gb GTX 560Ti Phantom 2 / 32 Gb DDR3 / Intel x25-M 160 Gb SSD and 2 x 1,5 Tb HDDs
_Lazarus_
2 x Intel Xeon MP Gallatin 3,00 Ghz with 4 MB cache / Zotac 512Mb GT430 / 12 Gb DDR266 ECC / 4 x Maxtor Atlas 146GB 10K V U320

Not only slow but crappy-looking and crappy-performing, too. I have yet to see a graphical Java app that doesn't look like dogpoop and leave little pieces of itself all over the screen. They should rebrand it to Junkva.

Which brings up something that's been bothering me for twenty years : if everybody is going to fall for the Pareto crap, then how come all the "smart" work that we are now supposed to be doing is such garbage ? Most of those apps referenced by [[C]-[E]] are done at universities !! Students, with all the time on Earth in their hands. Teachers, who are supposed to be guiding young minds ! And what do they produce, our Best of the Best, our 20% of the 20% that produces 80% of the profit ?

In a Chaucerian word ? Shitte. Absolute fricking useless garbage. Bowls and bowls of odiferous pigge shitte. And they are proud of it !

This ain't gonna work out well. I may not be Cassandra but even a cranky old bastard can see that convincing yourself that you're the smart 20% and deserve the big bucks while actually producing trash .... Dark Ages, here we come !

(Alright, I'll post what I started to tap out earlier and then thought better of.)

The CompSci faculty may not be involved in these projects, so the various CompSci fads may not be in play. At the same time people working on grant money are likely going to have to stretch it as far as possible - you may or may not be getting the top candidates who know they have other options.

For a long time now they've been teaching Java in Programming For Poets and CompSci 101 - I'm pretty far removed from the ivory tower to know if that's still the case, maybe they've shifted to C# or Haskell by now. But follow that through - any budding computational biologists are being taught Java, certainly not Tcl/Tk. And anybody likely to be hired as cheap student labor will know Java, and who knows what else. And you don't have to be a CompSci Ph.D to realize that somebody new will have to handle the code maintenance down the road, so if you try to convince them to write in some other language they're still likely to gravitate back to Java...

So I'm not surprised to hear that a lot of the projects (or front-ends) are written in Java. Or that they may not cleave to the highest standards of programming purity.

_________________
Then? ... Now? +MLA ... Other: DEC Sun, DG AViiON, NeXT

There. Fixed that for you


The place is in deep shit, smj. If TickleTock is too difficult for college students writing "we'll do the brain work, the lowlife chinks can do the dirty grunt work !" molecular biology applications, then the US is effing cooked. Stick a fork in it. Da poow widdow babies, might actually have to learn something ! And those meany mean teachers, no time to insist on any quality standards, they've got grants to pursue !

Here's a hint : the Chinese students coming to the US in droves now ? They aren't the smart ones. They are the lazy children of rich peasants (and shanghainese) who failed their gaokao, so the only option with status left to them is go overseas.

Toast, smj. Toast. Dry; no butter, cinammon or sugar. You guys are going to be lucky to be eating the crumbs on the floor if you don't wake up.

Too be fair, there is a lot of "Whadya want for nuthin" in the biology software market. It's one of the nichiest niches out there. Almost all of the free stuff is something that a biology grad student (not a comp sci grad student) banged out with a little perl or Java that he learned on the side. If you want clean, well maintained interfaces, be prepared to pay hundreds of dollars, at a minimum. Most of the nicer desktop packages start at around $1,500, which feels like a lot for an academic, but really is not unreasonable at all for a specialized, professional level tool.

It's one of the things that is really perverse in the academic world: grad students are really just two steps away from slave labor. Why pay $1,500 for a tool that will let a grad student focus on their research when you can save a few bucks by having them spend a few months reinventing the abacus? The professor actually ends up spending more on labor, time, and indirect costs than he would have spent just buying the tool, the student doesn't really learn anything important about their field of study, but it's hard to get away from the penny-wise/pound-foolish attitude that grad students are "free." Who cares if stuff like that adds a year or three to a PhD? ...and people wonder why bright kids go for MBAs instead of science/engineering degrees...

Thanks for the links! Those are definitely some of the more amazing programs I've seen in recent memory...

_________________
Choosing stones, big enough to drag me down...

You can also try Pymol or VMD. The last one is better and, since is CUDA capable, much faster .

http://www.pymol.org/

http://www.ks.uiuc.edu/Research/vmd/

_________________
_Betty Blue_
R12000A 400 Mhz; 1 Gb RAM; 72 Gb 15K HDD; IRIX 6.5.29
CrystalEyes; Dial Box; O2Cam "ZEYE"; external Toshiba SD-M1711 DVD-ROM; Octane speakers;
Lock bar; SGI microphone.
Mods: PSU Noctua fan; internal Toshiba SD-M1401 DVD-ROM; Adaptec AIC-7880P SCSI card.

_REKIEM_I7_
Seasonic X 1250W PSU / Intel I7 2600k 4 x 5,00 Ghz / 2 x Gainward 2Gb GTX 560Ti Phantom 2 / 32 Gb DDR3 / Intel x25-M 160 Gb SSD and 2 x 1,5 Tb HDDs
_Lazarus_
2 x Intel Xeon MP Gallatin 3,00 Ghz with 4 MB cache / Zotac 512Mb GT430 / 12 Gb DDR266 ECC / 4 x Maxtor Atlas 146GB 10K V U320

This is where I am at right now. I am new to SGI and have found several nice machines that might as well be paintings because I can hardly use them. That said, like rusti I enjoy getting the hardware to work. I enjoy searching websites and forums, scrounging ebay and just plain beating my head against these things. Nothing beats the feeling of sitting back and looking at the just repaired machine, even if I can't do anything with it yet.

_________________
(X3) (Fixed!) (x2) (X3)