This may just be my misunderstanding, but the way I understood it was that the OOM killer takes effect when there is no more memory available at all. You say "The worst that should happen is that it gets suspended with all of its pages taken away." but I have to wonder what the process is going to do when it starts up and has no pages - it'll crash instantly, so why not kill it, right?
Or did you mean that "the process's pages will be swapped?" Even if you did mean that, my understanding is that the OOM killer only takes effect when there is no memory space left - including swap. In this scenario, there isn't much to be done should the system need more memory to continue - you either kernel panic or you find some process to kill and kill it. In an extreme circumstance like OOM on a kernel alloc, I see nothing wrong with deciding to kill a process. I really don't see how "suspending" on a process solves memory issues since it still needs its pages somewhere...
My understanding was that the idea behind the OOM killer was to prevent the kernel from panicing and instead leave a working system which needs to have its memory problems worked out. I could be wrong since I haven't really looked into the OOM killer and when it's invoked.
Unfortunately, none of these are operating systems - they are all software packages and quite distinct from the OS to most people (er, except for WinTV, which is a hardware card). I seriously doubt "Lindows" has a leg to stand on. (I'd have named it something like "WinOnLin" or something else that gets the idea that it's a Windows emulator running on Linux, Lindows is a pretty dumb name... Although I suppose they could argue Lindows = LINux + WinDOWS, but I doubt that'll fly...)
Since this was posted by an AC, I'd like to reiterate this point with an actual datapoint.
Windows XP ran fine on a PII 400 with 256MB RAM and 5GB hard drive space. With all the pretty and useless GUI options enabled. Now it was a little slow, but no worse than GNOME on X on my nVidia GeForce2 on my 800MHz Athlon. The only thing that really killed the usability was excessive use of alpha fade effects in certain scenarios (namely, selecting a rectangle of files Windows Explorer) that weren't hardware accellerated due to an older graphics card.
For most "every day" tasks, the PII 400 was fine - you could browse the web, listen to MP3s, and play older XP-compatible games (which, in most cases, is the same as a Win2K compatible game).
Bottom line is that XP is no worse than any other "modern" graphical OS - it's just made by Microsoft. Accept the fact that Windows XP is a decent operating system and far superior to the Win9x line and get back to using your Linux PC. To each their own, but bashing XP without actually using it is pretty foolish, especially because it does run at without noticable slowdown on any new PC and on most older PCs as well.
Unless your desktop is still a Pentium class machine, assuming that your computer has enough disk space and RAM, Windows XP is a decent operating system. If you're going to bash it, bash it on the potential Digital Rights Management that was supposed to be introduced in XP, or on the product activation, or on any other Microsoft expansionist move. Bashing it for being slow is mostly just uninformed.
It's highly unlikely Slashdot actually records that information, but for the record, when I posted a mirror of some KDE screen shots, I recorded stats based on the browsers/OS used. (Server's down right now - it'll be back up when I return to college...)
I'd like to point out that the stats are completely and totally statistcally useless as A) they aren't random (arguably, they're based on the whole population, though - where that population is "people who wanted to see KDE3.0 screenshots" which is definately not the same thing as "people who view Slashdot"), and B) there are other factors that cannot be calculated, such as browsers that flat out lie (ie, report themselves as MSIE instead of Opera to get around broken websites), browsers that I don't recognize, people who use Linux as their home desktop but were viewing the images through work, and undoubtably other factors I can't think of.
Since Slashdot considers the table "lame" I'll just post the link to the journal article. For those who don't want to read through the post (the table's the second table, first is browser), the values most people want to know are 26.8% of the users reported themselves as using Linux, while 4.3% used another Unix (or were an X11 client...). 68.3% used some variety of Windows.
I've got more detailed information about the individual breakdown at home, but I'm at work... if anyone's interested, reply, and I'll see what I can come up with.
You're missing some of the backstory by not being from WPI...
The tree is located in the CCC shop (CCC = College Computer Center), which is the place A) where WPI computers that break are repaired (I think student computers can be brought there if they were bought via a WPI sponsored vendor, but I'm not sure...) and B) where Network Operations is located (making this slightly funnier IMHO).
For the past couple of years, there has been a tree in the CCC shop which is decorated with computer parts. This year, Paul Reitchel and Chuck Anderson decided to make those parts actually work. So they powered up the board with a power supply at the bottom of the tree and set it up to play Christmas songs via a web interface (although the speakers were, while not off, turned down to inaudiable when I swung by the shop - it was already popular enough around campus to drive everyone nuts, which doesn't surprise me).
Also, a "Christmas" tree has ornaments, lights, tinsel...christmasy things...
It does have lights, and, again, it's in the CCC shop, so it's decorated with hardware mostly as a joke. (Plus, if Chuck gets around to X10-ing the tree lights, it'll really be "Linux powered" - I think the fact that it plays music via the board is makes it "powered" enough... How would you use Linux to power a tree? Make it rotate or something?)
And if you want to see something really funny, take a look at the I2 link upstream/downstream graphs. You can see when the story was posted on the daily graph at about 20:00 EST.
(Plus the idea of NetOps taking up more than the maximum allowed bandwidth per system is a rather amusing idea to a WPI student...)
The script checks for browsers in roughly the order of Konqueror, Opera, MSIE, misc browsers that include Mozilla/4.0 (compatible;, and then Mozilla/Netscape. But yeah, if you actually set the user agent to start exactly "Mozilla/4.0 [en]..." (or any other non-whitespace non-"]" (ie, User Agent =~/Mozilla\/4.0 \[[\S\]]+\]/)) it'll count towards Netscape 4. If you leave off the language code you get dumped into "Other." Likewise, Mozilla/4.0 (compatible, MSIE 5.0, Windows NT 5.0) will get you counted as Internet Explorer on Windows 2000.
So if you want to get most pages to decide you're an OK browser, do what Opera does and change it to something like "Mozilla/4.0 (compatible; MSIE 5.0; Linux 2.4.14 i686) Opera 5.0 [en]" which will count towards an Opera hit via my script and make web developers decide that maybe there is a reason to pay attention to non-IE browsers...
(The script was based off code originally to check which browsers were being used to view a webpage and basically made the decision that the site would be "optimized" for MSIE and the other 5% could just deal.)
Then you'll be detected as the browser you're actually using instead of just counting as yet another IE hit.
Unless you do something truely weird like "Mozilla/5.0 (compatible; MSIE 5.5; oops, wrong! it's:) Mozilla/5.0 Galeon/0.12.5 (Linux i686; U;) Gecko/20011012" (counts as Galeon) or "Mozilla/6.66 (compatible; MSIE 7.23; Windows GPF 5.0)" or even "Mozilla/6.0 [en] (compatible; MSIE 9.81; Sinclair ZX81 BASIC)" (both of which were counted as an MSIE hit)...) or probably the best "RubeBrowser/42.0 (C64 Geos; Liquid Helium Cooled)".
The 9 hits by "QuickTime (qtver=5.0.8;os=Windows NT 5.2)" I have to wonder about...
A little less than half the Slashdotters. (Maybe. This data is a non-random sample, since I can't really do a random sample, although someone with access to Slashdot's servers could...)
Read my journal entry about how I got this data, or just look at the table (that cannot be formatted properly because the lameness filter is the most useless piece of crap that Slashdot has ever forced upon its readers - I'm glad you guys are all about free speech online!! - so use the linked journal where the formatting was accepted and don't forget to continously annoy CmdrTaco about this annoying "feature" to protect us from the oh-so-evil trolls):
Browser Actually Used By Slashdotters
Galeon: 1511 (3.00%) iCab 9 (0.02%) Konqueror 4149 (8.25%) Lynx 6 (0.01%) Internet Explorer 24885 (49.47%) Mozilla 9340 (18.57%) Netscape 3756 (7.47%) OmniWeb 190 (0.38%) Opera 3267 (6.50%) Other 3187 (6.34%)
Note: Other contains browsers whose User-Agents could not be parsed. It may contain valid browsers, but for the most part is either badly formed User-Agent strings or unknown User Agents.
It has to be noted again that this data is not statistically accurate: it was taken directly off of hits, and is biased towards browsers that automatically download images (in other words, every hit counted - the values didn't take into account which hits were hits to the images linked to on the page).
Also, some other people decided to... uh, borrow... the mirror and so some of the links come from other sources that aren't Slashdot. I forget if I filtered those or not, but...
If anyone's interested, I suppose I could try and fix up the Perl scripts used to calculate that data. I have some pretty pie charts on my harddrive that I could put up somewhere too, although they are for the most part useless...
Also, make it so you, the user can resize the font. N[o]t sure how it works, but I've seen my share of pages where moving the font size up and down doesn't work at all. People with poor eyesight will be thankful.
Someone else mentioned it but said they weren't sure, so I'll explain it more exactly:
Sites that specify font sizes as something concrete (ie, points or pixels) cause most browsers (read: Internet Explorer) to fail to resize the text. My webpage lays everything out via CSS (as in, no tables, but a menu on the left and content on the right). (No comments on usability please, I use it for myself and I can use the thing - actually, I think the base layout works fairly well, but my color scheme probably will piss off people who aren't me:). Also, a word of warning: if your browser does not support CSS and HTML4.0, the page will look funky. It'll render quasi-alright in Lynx (the menu currently comes first - maybe I should fix that) and it'll layout (almost) properly in Netscape 4, but the fonts will be all screwed up. Ignoring an off-by-one bug in Moz, it renders fine in both Mozilla and IE.) I use points to specify the layout since I would hope that specifying it in points would allow a browser to scale the font and don't want to play the "match layout with percent font size" game that I'd have to play otherwise.
Specifying font sizes as percents allow IE to properly scale the font. Specifying the size as points or pixels causes IE to keep the text size static when "zoomed." (Mozilla scales the font reguardless of point/percent, but I haven't tried pixel - my guess is that it would scale it as well.)
Personally, my view is that when text is "zoomed" the point to pixel conversion should be scaled up - IE doesn't scale the points and Mozilla only scales the font sizes. (Again, if you look at my webpage, I think that when I scale the font size up, the layout should scale along with the new font size since the font size and the layout are both specified in points. It doesn't scale at all in IE and Mozilla just makes the glyphs larger/smaller.)
Arguably, the fact that IE doesn't scale the font sizes larger is a bug in IE and not an issue with the web developers. But YMMV, I suppose it would make sense to redo the webpage to work around bugs in the most popular browser, especially one that makes pages illegible in it.
Hey, if you're still looking for work, are you willing to work for free in a crappy movie [no web page - yet] for a crappy movie contest?
We still need a camera, a script, actors, props, costumes, sets, and a budget, but we've got a basic story line going... If you're interested, then you are insane.
Trying to discount the fact that Bruce Campbell's interview was related third-person...
I wouldn't do that if I were you. That's would be like saying "Trying to discount the fact that the car is on fire, the previous car that wasn't engulfed in flames looks like it'd be a safer ride."
Seriously, the whole tone of this interview is ruined because chrisd doesn't have a tape recorder and didn't think to try and schedule a better time for the interview. I mean really, Wil answered using his own writing style, Bruce phoned in and had Chris butcher whatever he said in a half-assed sort of transcription.
[Chris: For future reference, should something like this come up again, tell the interviewee to hold off for a while until you can set up a way to record the call and then properly transcribe the interview. At least try and edit the results so that the mix-and-match "Bruce thinks" and direct quotes reads a little better. This reads like notes I take in class on my iPaq - I don't have time to write down every word, but I write down what I think it's important to know for the class. In the case of an interview, it really should be recorded so an appropriate transcription can be made - every word the interviewee speaks is important to the audience.]
The better interview was the one where the interviewee wrote his own answers and didn't rely on a third party to get the message across. It's not suprising that Wil's interview sounded better.
(Although this interview at least delineated the end of the question and the start of the answer...)
I think what they were suggesting is that you had done something seemingly boneheaded like:
User-Agent: * Disallow:/
Think about it - how many people do you think are out there with a half-clue who decide that they want to prevent evil robots from indexing their site without realizing that they therefore won't wind up on search engines? Apparently Google seems to have run into this situation and now e-mails webmasters who have potentially accidently blocked all robots from indexing their pages.
Now there may be a valid reason to completely block your site from all robots. But think about how pointless it really is - how many webmasters really want to drive away search engines? Most people want to show up on search engines, especially people whose site shows up as a domain (ie, http://slashdot.org/ as opposed to http://www.wherever.edu/~they/started/).
Seriously, why did you block the entire domain from web crawlers? While there definately are good reasons, it seems sensible for Google to send a "are you really sure you want to do that" message, especially since the linked "spam" was sent to someone who apparently had four domains they had blocked off from search engines. This sounds like something that an amatuer webmaster may have accidently done without thinking about the consequences. In which case the e-mail makes sense: "Did you really mean to do that? If so, ignore this message - if not, here's a way to fix it."
I really think you're overreacting to a fairly innocent e-mail.
Three posts in one day from Alan Cox? Are you really that bored or has someone hijacked your account? Or maybe you're moving to karma whoring?:P
(Actually, it would seem that Alan is starting to take an active position on Internet / Digital rights - and that he's becoming more of an obsessive/. reader than I am!)
Well, the blinking lights on my hub are still up, and I seem to have stopped at around 31936 hits. Er, 31943 hits. Make that 31951 hits. Anyway, I don't think I hit the limit:)
Sometime tomorrow I'll write a Perl script to parse hits and figure out how much bandwidth I wasted. Plus other fun stats like what web browsers people were using and stuff like that (yes, I keep track:)).
I'll post it as a journal entry to anyone interested. (Plus stick more detailed reports somewhere in the offtopic section of cstrike.)
Re:What makes a good IDE, aka: Netbeans is real cl
on
Java IDEs?
·
· Score: 1
This is on win2k, so I'm sure it'd probably work even better on a Unix.;-)
My experience with Java is that anything AWT is far faster on Windows than on any UNIX for the same box. (IE, if I run an AWT app when booted to Linux, it'll run slower than when the machine is booted into Windows - including 9x.)
Likewise, assuming you can find an equivilent Solaris box, AWT apps will run faster under Windows than Solaris. It's just that AWT apps (especially Swing apps) take so much time up fiddling with drawing code and translating between architectures that the Windows code winds up being faster.
(I think this also might be due to the inherent speed hit the X11 architecture forces too. Face it, calling a function directly is faster than sending a command for that function across a pipe - at the very least, you add an encode -> decode step. Be interesting to benchmark Win32 GTK+ against X11 GTK+. 'Course, then again, the drivers would need to be made the same, and...)
Once you move from GUIs into server-land, then yes, UNIX's strengths help Java. But when it comes to using Java for a GUI, Windows Java is faster (assuming that everything is user-driven - i.e., no actions take place until the user initiates them, i.e., a text editor).
Anyways--linux is doing fine. Anybody running XP on a p2 266?
Actually, yes. I've used Windows XP on a P2 450 before and it worked quite nicely. It's reasonable to suspect it will work with a 266, although it may need to be moved out of true color mode.
Anyway, my Dad sent me the following about someone who truly decided to shoot themselves in the foot, but anyway:
It seems that he [a course instructor teaching my Dad's group] wondered if you could install it on a system that was less than the minimum requirements published by Microsoft.
He got out a very old system - principle use: doorstop - and tried to install Windows XP RC2 on it. Here are the results:
1) System: 386SX16 with 32Mb memory
2) Install was successful. But it took 5-1/2 DAYS!
3) Time to boot up: 8 hours
4) Time to start up Microsoft Word: 45 minutes
Not a very productive configuration but it actually ran (walked? crawled? even slower?)
First of all,
this is a great example of how not to write a Palm version of the site, and here's why. Offline readers depend on "link-depth" to traverse a site. However, their Palm version breaks each story into a random number of small chunks. So, you can't just page-down to read a long story or a bunch of comments- you have to click on lots and lots of links. A real pain. Lots of small links makes sense on a slow online connection, but it's awful when you have more bandwidth available, as your desktop PC or an offline browser.
Additionally, it's restricted to 10 comments, not a threshold. That's boring. I'm sitting here in Jiffy Lube picking my nose, I wanna read some funny trolls and flamewars!
Finally, using/. in "light" mode doesn't work either. There are too many useless links on the front page. I don't care about the advertising or the FAQ or all the other stuff: I want the stories and the comments. Basically, the readers I use so far have no way to "prune" sections of the tree you don't care about. This causes the site to be gigantic and not fit into the paltry 8MB of your typical handheld, or, it fits, but it so big as to detract from its usefulness.
Finally, someone did the right thing: AvantSlash takes the page, filters out all the crap you don't care about, and doesn't break it up into a thousand chunks so it's readable.
In order to make that usable, I'd have to pump my link depth to something like 4 in order to read the stories. Plus, for the first time in months, slashdot.org has stopped serving 403's to sync.avantgo.com, which basically killed it's usefulnes... (It was one of the first sites I tried to sync to my iPaq via AvantGo, and until today, everytime I tried, I'd get access denied errors reading it when I tried to sync.)
Example of a good balance of storyline, action, effects and ending for Sci-fi movies: Terminator II. You can have an hollywood ending without spoiling the entire movie...
Ironically enough, Terminator II's originally scripted ending was terrible: it took place 30 years in the future (I think) and had what's-her-name O'Connor recording her memoires as an old lady in a happy-fancy future. (Next to the playground where all the children are playing and got blown up in the nuclear strike shots - except that the buildings are drawn in this cheasy-futuristic style.)
Thank God (or your deity/spirits of choice) that they came to their senses and replaced the ending with the final one that came out on film...
Yeah, I know: with VA Linux listed as a 1 out of 10 as a "good purchase" via StockScouter, I know I feel confident in investing my future in a company that has no known future direction.
VA Linux Systems, Inc., a micro-cap value company in the technology sector, is expected to
significantly underperform the market over the next six months with very high risk.
They recognized it too!
Seriously, though, does/. have a backout plan? I hope some other company thinks it's worthwhile, but it's fairly obvious that VA $WHATEVER isn't going to be around much longer.
I believe this also spells impending trouble for Kuro5hin, Freshmeat, NewsForge, obvious SourceForge (good thing I'm not hosting anything there), and I'm sure other people can list other things.
I wonder what will happen to all these sites when (or if) VA goes under? Will they be liquidated the highest bidder? I wonder what happens to code on SourceForge? (I wonder if a Linux distro would be interested in picking up SourceForge - seems like it could only help them...)
Many things may happen to the Open Source community should VA die...
I wonder who many other people didn't know what an RFP was? I'll bet most of us have never been through the process of sending out RFPs. (Ironically, I was, but only very briefly, and forgot the meaning of the term...)
An RFP stands for Request For Proposals and are sent out when a company wants other companies to complete something for it. (It's a request for proposals to complete a certain task - also known as a Request for Contracts.)
As a general reference to Open Source packages, I'd either send RFCs out to a company that deals with them (ie, RedHat is known to do work with several open source packages, so for an RFP involving a piece of software they package, they would be a good choice to receive an RFP).
In the case of Open Source projects that happen not to have a company involved with them, you could try and look for general purpose contractors, I suppose, or you could look into doing the work internally. You might also look into general software contracters.
It may also be possible to contract with the people working directly on the OSS project, so you should definately try and contact people on the project list (if to be polite, for no other reason) but be aware that most OSS developers are either A) already employed, or B) students. (The "project list" in this case is either a developer list if the documentation included with the source distribution includes such information or the list of people in the AUTHORS file that many Open Source projects include - assuming it's up-to-date.) In either case, it's highly probable that the active maintainers will not be able to send back a useful proposal, and your best bet is looking for outside contracters.
(This of course could open up an interesting buisness plan - a software house that specializes in creating contracts to do work on Open Source software...)
My first experiences with Mandrake were not good. It destroyed my Windows partition without warning. (fortunately I had a backup, but this was hardly a user-friendly thing to do).
I can beat you on that one. I was installing Mandrake on a test box to see how well WinXP RC2 liked dual-booting, and Mandrake deleted the partition table. When you try and fix that via fdisk and it bombs because the partition table is so badly corrupted, you've moved from "not user-friendly" to downright evil.
(Actually, it didn't delete the partition table. It wrote it in such a way that the extended partition overlapped with the primary partion. Which is more than enough to throw Partition Magic 6.0 and fdisk for a loop, apparently - enough so that the only solution was to rewrite the table for scratch. Effectively deleting it. Some how, the Linux kernel didn't mind the state of the partition table, though...)
So, yeah, Mandrake is perfectly capable of screwing up your system and completely screwing up partitions.
(From now on, I always partition a new system first and then install Mandrake...)
Slashdot Mirror
Or did you mean that "the process's pages will be swapped?" Even if you did mean that, my understanding is that the OOM killer only takes effect when there is no memory space left - including swap. In this scenario, there isn't much to be done should the system need more memory to continue - you either kernel panic or you find some process to kill and kill it. In an extreme circumstance like OOM on a kernel alloc, I see nothing wrong with deciding to kill a process. I really don't see how "suspending" on a process solves memory issues since it still needs its pages somewhere...
My understanding was that the idea behind the OOM killer was to prevent the kernel from panicing and instead leave a working system which needs to have its memory problems worked out. I could be wrong since I haven't really looked into the OOM killer and when it's invoked.
- WinAMP
- WinZip
- WinRAR
- WindowBlinds (well, it's not "Windows"...)
- WinTV
Of course, all of these are basically software packages that run on Windows. (Or hardware...)Then of course there's X-Windowing System...
Unfortunately, none of these are operating systems - they are all software packages and quite distinct from the OS to most people (er, except for WinTV, which is a hardware card). I seriously doubt "Lindows" has a leg to stand on. (I'd have named it something like "WinOnLin" or something else that gets the idea that it's a Windows emulator running on Linux, Lindows is a pretty dumb name... Although I suppose they could argue Lindows = LINux + WinDOWS, but I doubt that'll fly...)
Windows XP ran fine on a PII 400 with 256MB RAM and 5GB hard drive space. With all the pretty and useless GUI options enabled. Now it was a little slow, but no worse than GNOME on X on my nVidia GeForce2 on my 800MHz Athlon. The only thing that really killed the usability was excessive use of alpha fade effects in certain scenarios (namely, selecting a rectangle of files Windows Explorer) that weren't hardware accellerated due to an older graphics card.
For most "every day" tasks, the PII 400 was fine - you could browse the web, listen to MP3s, and play older XP-compatible games (which, in most cases, is the same as a Win2K compatible game).
Bottom line is that XP is no worse than any other "modern" graphical OS - it's just made by Microsoft. Accept the fact that Windows XP is a decent operating system and far superior to the Win9x line and get back to using your Linux PC. To each their own, but bashing XP without actually using it is pretty foolish, especially because it does run at without noticable slowdown on any new PC and on most older PCs as well.
Unless your desktop is still a Pentium class machine, assuming that your computer has enough disk space and RAM, Windows XP is a decent operating system. If you're going to bash it, bash it on the potential Digital Rights Management that was supposed to be introduced in XP, or on the product activation, or on any other Microsoft expansionist move. Bashing it for being slow is mostly just uninformed.
If I SSH to my school, which does not cache domain names, I get the following:
Server: non-caching.name.server
Address: 192.168.1.1
*** non-caching.name.server can't find www.theregister.co.uk: Non-existent host/domain
Whereas if I run the same command here, I get:
Server: caching.name.server
Address: 192.168.1.2
Non-authoritative answer:
Name: www.theregister.co.uk
Address: 213.40.196.64
So those without it cached can read it via http://213.40.196.64/ or you can just add it to /etc/hosts or %SYSTEMROOT%\system32\drivers\etc\hosts
I'd like to point out that the stats are completely and totally statistcally useless as A) they aren't random (arguably, they're based on the whole population, though - where that population is "people who wanted to see KDE3.0 screenshots" which is definately not the same thing as "people who view Slashdot"), and B) there are other factors that cannot be calculated, such as browsers that flat out lie (ie, report themselves as MSIE instead of Opera to get around broken websites), browsers that I don't recognize, people who use Linux as their home desktop but were viewing the images through work, and undoubtably other factors I can't think of.
Since Slashdot considers the table "lame" I'll just post the link to the journal article. For those who don't want to read through the post (the table's the second table, first is browser), the values most people want to know are 26.8% of the users reported themselves as using Linux, while 4.3% used another Unix (or were an X11 client...). 68.3% used some variety of Windows.
I've got more detailed information about the individual breakdown at home, but I'm at work... if anyone's interested, reply, and I'll see what I can come up with.
The tree is located in the CCC shop (CCC = College Computer Center), which is the place A) where WPI computers that break are repaired (I think student computers can be brought there if they were bought via a WPI sponsored vendor, but I'm not sure...) and B) where Network Operations is located (making this slightly funnier IMHO).
For the past couple of years, there has been a tree in the CCC shop which is decorated with computer parts. This year, Paul Reitchel and Chuck Anderson decided to make those parts actually work. So they powered up the board with a power supply at the bottom of the tree and set it up to play Christmas songs via a web interface (although the speakers were, while not off, turned down to inaudiable when I swung by the shop - it was already popular enough around campus to drive everyone nuts, which doesn't surprise me).
Also, a "Christmas" tree has ornaments, lights, tinsel...christmasy things...
It does have lights, and, again, it's in the CCC shop, so it's decorated with hardware mostly as a joke. (Plus, if Chuck gets around to X10-ing the tree lights, it'll really be "Linux powered" - I think the fact that it plays music via the board is makes it "powered" enough... How would you use Linux to power a tree? Make it rotate or something?)
And if you want to see something really funny, take a look at the I2 link upstream/downstream graphs. You can see when the story was posted on the daily graph at about 20:00 EST.
(Plus the idea of NetOps taking up more than the maximum allowed bandwidth per system is a rather amusing idea to a WPI student...)
So if you want to get most pages to decide you're an OK browser, do what Opera does and change it to something like "Mozilla/4.0 (compatible; MSIE 5.0; Linux 2.4.14 i686) Opera 5.0 [en]" which will count towards an Opera hit via my script and make web developers decide that maybe there is a reason to pay attention to non-IE browsers...
(The script was based off code originally to check which browsers were being used to view a webpage and basically made the decision that the site would be "optimized" for MSIE and the other 5% could just deal.)
Then you'll be detected as the browser you're actually using instead of just counting as yet another IE hit.
Unless you do something truely weird like "Mozilla/5.0 (compatible; MSIE 5.5; oops, wrong! it's:) Mozilla/5.0 Galeon/0.12.5 (Linux i686; U;) Gecko/20011012" (counts as Galeon) or "Mozilla/6.66 (compatible; MSIE 7.23; Windows GPF 5.0)" or even "Mozilla/6.0 [en] (compatible; MSIE 9.81; Sinclair ZX81 BASIC)" (both of which were counted as an MSIE hit)...) or probably the best "RubeBrowser/42.0 (C64 Geos; Liquid Helium Cooled)".
The 9 hits by "QuickTime (qtver=5.0.8;os=Windows NT 5.2)" I have to wonder about...
Read my journal entry about how I got this data, or just look at the table (that cannot be formatted properly because the lameness filter is the most useless piece of crap that Slashdot has ever forced upon its readers - I'm glad you guys are all about free speech online!! - so use the linked journal where the formatting was accepted and don't forget to continously annoy CmdrTaco about this annoying "feature" to protect us from the oh-so-evil trolls):
Browser Actually Used By Slashdotters
Galeon: 1511 (3.00%)
iCab 9 (0.02%)
Konqueror 4149 (8.25%)
Lynx 6 (0.01%)
Internet Explorer 24885 (49.47%)
Mozilla 9340 (18.57%)
Netscape 3756 (7.47%)
OmniWeb 190 (0.38%)
Opera 3267 (6.50%)
Other 3187 (6.34%)
Note: Other contains browsers whose User-Agents could not be parsed. It may contain valid browsers, but for the most part is either badly formed User-Agent strings or unknown User Agents.
It has to be noted again that this data is not statistically accurate: it was taken directly off of hits, and is biased towards browsers that automatically download images (in other words, every hit counted - the values didn't take into account which hits were hits to the images linked to on the page).
Also, some other people decided to ... uh, borrow ... the mirror and so some of the links come from other sources that aren't Slashdot. I forget if I filtered those or not, but...
If anyone's interested, I suppose I could try and fix up the Perl scripts used to calculate that data. I have some pretty pie charts on my harddrive that I could put up somewhere too, although they are for the most part useless...
Audio file of said quote off my web server.
Someone else mentioned it but said they weren't sure, so I'll explain it more exactly:
Sites that specify font sizes as something concrete (ie, points or pixels) cause most browsers (read: Internet Explorer) to fail to resize the text. My webpage lays everything out via CSS (as in, no tables, but a menu on the left and content on the right). (No comments on usability please, I use it for myself and I can use the thing - actually, I think the base layout works fairly well, but my color scheme probably will piss off people who aren't me :). Also, a word of warning: if your browser does not support CSS and HTML4.0, the page will look funky. It'll render quasi-alright in Lynx (the menu currently comes first - maybe I should fix that) and it'll layout (almost) properly in Netscape 4, but the fonts will be all screwed up. Ignoring an off-by-one bug in Moz, it renders fine in both Mozilla and IE.) I use points to specify the layout since I would hope that specifying it in points would allow a browser to scale the font and don't want to play the "match layout with percent font size" game that I'd have to play otherwise.
Specifying font sizes as percents allow IE to properly scale the font. Specifying the size as points or pixels causes IE to keep the text size static when "zoomed." (Mozilla scales the font reguardless of point/percent, but I haven't tried pixel - my guess is that it would scale it as well.)
Personally, my view is that when text is "zoomed" the point to pixel conversion should be scaled up - IE doesn't scale the points and Mozilla only scales the font sizes. (Again, if you look at my webpage, I think that when I scale the font size up, the layout should scale along with the new font size since the font size and the layout are both specified in points. It doesn't scale at all in IE and Mozilla just makes the glyphs larger/smaller.)
Arguably, the fact that IE doesn't scale the font sizes larger is a bug in IE and not an issue with the web developers. But YMMV, I suppose it would make sense to redo the webpage to work around bugs in the most popular browser, especially one that makes pages illegible in it.
We still need a camera, a script, actors, props, costumes, sets, and a budget, but we've got a basic story line going... If you're interested, then you are insane.
I wouldn't do that if I were you. That's would be like saying "Trying to discount the fact that the car is on fire, the previous car that wasn't engulfed in flames looks like it'd be a safer ride."
Seriously, the whole tone of this interview is ruined because chrisd doesn't have a tape recorder and didn't think to try and schedule a better time for the interview. I mean really, Wil answered using his own writing style, Bruce phoned in and had Chris butcher whatever he said in a half-assed sort of transcription.
[Chris: For future reference, should something like this come up again, tell the interviewee to hold off for a while until you can set up a way to record the call and then properly transcribe the interview. At least try and edit the results so that the mix-and-match "Bruce thinks" and direct quotes reads a little better. This reads like notes I take in class on my iPaq - I don't have time to write down every word, but I write down what I think it's important to know for the class. In the case of an interview, it really should be recorded so an appropriate transcription can be made - every word the interviewee speaks is important to the audience.]
The better interview was the one where the interviewee wrote his own answers and didn't rely on a third party to get the message across. It's not suprising that Wil's interview sounded better.
(Although this interview at least delineated the end of the question and the start of the answer...)
User-Agent: * /
Disallow:
Think about it - how many people do you think are out there with a half-clue who decide that they want to prevent evil robots from indexing their site without realizing that they therefore won't wind up on search engines? Apparently Google seems to have run into this situation and now e-mails webmasters who have potentially accidently blocked all robots from indexing their pages.
Now there may be a valid reason to completely block your site from all robots. But think about how pointless it really is - how many webmasters really want to drive away search engines? Most people want to show up on search engines, especially people whose site shows up as a domain (ie, http://slashdot.org/ as opposed to http://www.wherever.edu/~they/started/).
Seriously, why did you block the entire domain from web crawlers? While there definately are good reasons, it seems sensible for Google to send a "are you really sure you want to do that" message, especially since the linked "spam" was sent to someone who apparently had four domains they had blocked off from search engines. This sounds like something that an amatuer webmaster may have accidently done without thinking about the consequences. In which case the e-mail makes sense: "Did you really mean to do that? If so, ignore this message - if not, here's a way to fix it."
I really think you're overreacting to a fairly innocent e-mail.
(Actually, it would seem that Alan is starting to take an active position on Internet / Digital rights - and that he's becoming more of an obsessive /. reader than I am!)
Go wild.
Sometime tomorrow I'll write a Perl script to parse hits and figure out how much bandwidth I wasted. Plus other fun stats like what web browsers people were using and stuff like that (yes, I keep track :)).
I'll post it as a journal entry to anyone interested. (Plus stick more detailed reports somewhere in the offtopic section of cstrike.)
We'll see how long this box can last...
My experience with Java is that anything AWT is far faster on Windows than on any UNIX for the same box. (IE, if I run an AWT app when booted to Linux, it'll run slower than when the machine is booted into Windows - including 9x.)
Likewise, assuming you can find an equivilent Solaris box, AWT apps will run faster under Windows than Solaris. It's just that AWT apps (especially Swing apps) take so much time up fiddling with drawing code and translating between architectures that the Windows code winds up being faster.
(I think this also might be due to the inherent speed hit the X11 architecture forces too. Face it, calling a function directly is faster than sending a command for that function across a pipe - at the very least, you add an encode -> decode step. Be interesting to benchmark Win32 GTK+ against X11 GTK+. 'Course, then again, the drivers would need to be made the same, and...)
Once you move from GUIs into server-land, then yes, UNIX's strengths help Java. But when it comes to using Java for a GUI, Windows Java is faster (assuming that everything is user-driven - i.e., no actions take place until the user initiates them, i.e., a text editor).
Actually, yes. I've used Windows XP on a P2 450 before and it worked quite nicely. It's reasonable to suspect it will work with a 266, although it may need to be moved out of true color mode.
Anyway, my Dad sent me the following about someone who truly decided to shoot themselves in the foot, but anyway:
Reproduced for the terminaly lazy:
In order to make that usable, I'd have to pump my link depth to something like 4 in order to read the stories. Plus, for the first time in months, slashdot.org has stopped serving 403's to sync.avantgo.com, which basically killed it's usefulnes... (It was one of the first sites I tried to sync to my iPaq via AvantGo, and until today, everytime I tried, I'd get access denied errors reading it when I tried to sync.)
I don't see what difference "not for sale in MA" would do anyway... :)
(I get all my computer stuff in Nashua, NH - town of malls RIGHT NEXT to the MA-NH border... :))
Ironically enough, Terminator II's originally scripted ending was terrible: it took place 30 years in the future (I think) and had what's-her-name O'Connor recording her memoires as an old lady in a happy-fancy future. (Next to the playground where all the children are playing and got blown up in the nuclear strike shots - except that the buildings are drawn in this cheasy-futuristic style.)
Thank God (or your deity/spirits of choice) that they came to their senses and replaced the ending with the final one that came out on film...
They recognized it too!
Seriously, though, does /. have a backout plan? I hope some other company thinks it's worthwhile, but it's fairly obvious that VA $WHATEVER isn't going to be around much longer.
I believe this also spells impending trouble for Kuro5hin, Freshmeat, NewsForge, obvious SourceForge (good thing I'm not hosting anything there), and I'm sure other people can list other things.
I wonder what will happen to all these sites when (or if) VA goes under? Will they be liquidated the highest bidder? I wonder what happens to code on SourceForge? (I wonder if a Linux distro would be interested in picking up SourceForge - seems like it could only help them...)
Many things may happen to the Open Source community should VA die...
An RFP stands for Request For Proposals and are sent out when a company wants other companies to complete something for it. (It's a request for proposals to complete a certain task - also known as a Request for Contracts.)
As a general reference to Open Source packages, I'd either send RFCs out to a company that deals with them (ie, RedHat is known to do work with several open source packages, so for an RFP involving a piece of software they package, they would be a good choice to receive an RFP).
In the case of Open Source projects that happen not to have a company involved with them, you could try and look for general purpose contractors, I suppose, or you could look into doing the work internally. You might also look into general software contracters.
It may also be possible to contract with the people working directly on the OSS project, so you should definately try and contact people on the project list (if to be polite, for no other reason) but be aware that most OSS developers are either A) already employed, or B) students. (The "project list" in this case is either a developer list if the documentation included with the source distribution includes such information or the list of people in the AUTHORS file that many Open Source projects include - assuming it's up-to-date.) In either case, it's highly probable that the active maintainers will not be able to send back a useful proposal, and your best bet is looking for outside contracters.
(This of course could open up an interesting buisness plan - a software house that specializes in creating contracts to do work on Open Source software...)
I can beat you on that one. I was installing Mandrake on a test box to see how well WinXP RC2 liked dual-booting, and Mandrake deleted the partition table. When you try and fix that via fdisk and it bombs because the partition table is so badly corrupted, you've moved from "not user-friendly" to downright evil.
(Actually, it didn't delete the partition table. It wrote it in such a way that the extended partition overlapped with the primary partion. Which is more than enough to throw Partition Magic 6.0 and fdisk for a loop, apparently - enough so that the only solution was to rewrite the table for scratch. Effectively deleting it. Some how, the Linux kernel didn't mind the state of the partition table, though...)
So, yeah, Mandrake is perfectly capable of screwing up your system and completely screwing up partitions.
(From now on, I always partition a new system first and then install Mandrake...)