Heh, interesting they chose Mathematica. Of course it was easy to port: it already ran on multiple processors (x86 included of course) and a bunch of OSes (OSX, Win, Linux). The changes were probably just a bunch of #define's.
And even then it sounds better than it is. Yes Mathematica is a very large complex application with vast amounts of code. Most of that code is written in Mathematica itself however, and requires no porting whatsoever if you can get the Mathematica interpreter up and running.
Think of it as like porting Emacs: you only really have to get the elisp engine running, and most everything else will then instantly fall into place. Yes that takes some work, but vastly less work than it might at first look like based on the apparent complexity of the application.
It doesn't shock me too much that it only took 2 hours to port Mathematica. I mean, the API for OS X on Intel is probably exactly the same as for OS X on PPC. Probably only very, very small parts (if any at all) of Mathematica are written in assembly code. You fix those parts and anything that relies on specific processor behavior then do a recompile.
The majority of Mathematica is written in Mathematica. Porting Mathematica over is probably akin to porting Emacs: you get the elisp going, and you're pretty much done.
Sure there is some code to port over, but remember that Mathematica runs on Windows, MacOS X, Linux, and Solaris, and that the majority of the code is in Mathematica and doesn't need to be changed... I'm a little surprised it took 2 hours.
Who among us doesn't have files saved in an old version of, say, Word, which can no longer be read correctly in a newer version of Word?
Not really. I do have a lot of rather old TeX documents from an older version of TeX, some are 10 or 15 years old. All of them produce results unerringly identical to the old version. I fully expect to be able to use the TeX of 10 or 15 years in the future to still produce exactly the same output from the source. I expect similar things could be said about nroff and the like, but I don't use those, so I can't say for sure.
The really interesting thing is that I now use pdftex on the documents to get PDF output instead of DVI and Postscript. I'm sure if a new display format becomes similarly popular in the future there'll be a version of TeX to compile directly to the new format as well.
I don't fully understand people locking away their documents the way they do...
Jedidiah.
Re:This sounds normal
on
Safari vs. KHTML
·
· Score: 2, Interesting
I think this is pretty much the whole point: WebCore is a fork. There isn't much cooperation, or joint effort between WebCore and KHTML. Sure, some code manages to go both ways, but slowly but surely the two projects are diverging more and more. This isn't really a problem, the problem is expectations. The KDE people are pissed not so much at Apple, but rather at all the people that keep expecting changes in WebCore to come to KHTML. They are pissed with all the people that don't seem to understand that WebCore is a fork, and is now quite a separate project.
Go read some of the antiquated mailinglists for GNU/Emacs, I'm sure you'll find some equally pissed developers there complaining about people constantly asking when XEmacs changes will get merged into GNU/Emacs. That's all we're dealing with here.
Uh, yeah Bill, we've heard this promise before. I'm not holding my breath over any Microsoft promise that ends with "a thing of the past." The past keeps coming back to haunt you with Windows.
I think the thing to note is "without the users' knowledge" - I think we can still expect plenty of malware, viruses, and whatnot, they will simply be using social engineering attacks which, let's be frank, have proved remarkably effective. How many email viruses actually required user intervention to work? Building a system that is actually resistent to that is the real quest here.
The best solution to these sorts of problems is trusted computing and that comes in two flavours: the one provided by Microsoft and NGSCB, the one provided by TrustedSolaris, SELinux, TrustedBSD etc. The first locks down all the resources only giving access according to predefined rules: the general purpose computer becomes an appliance; it has a hardwired set of things that it can do. The second locks down resources only giving access according to an editable policy: if you really want to do something you can change the policy; you still have a general purpose computer.
The first flavour can, in theory, block the majority of social engineering attacks, but could become unusably constrained in so doing. The second flavour simply adds an extra layer to social engineering attacks: you have to convince the user to alter their policy as well as whatever steps were required for your malware. On the other hand, you can have a much tighter default policy.
Personally I like the second approach - I think SELinux looks excellent, and I am very keen to see OpenSolaris which includes TrustedSolaris. What the community needs to do is write/modify software to better respect the constraints that such systems allow, write good default policies, and create some good simple GUI tools for dealing with/editing security policies.
Adobe will have much less chance of making money on the Linux market, because the Gimp is preinstalled on so many machines. This breaks the 3rd party market on Linux machines.
Not really no. It breaks the 3rd party market for Debian and other distros that want to be 100% FOSS, but other distros will happily bundle proprietary 3rd party apps. Linspire, for instance, bundles a payware DVD player.
If Adobe wanted to get into the Linux market they can go and chat to Redhat and SuSE and Sun and Linspire and set up a deal to have Photoshop for Linux included in the distro. How that deal pans out is unknown - maybe it only gets included in RHEL but not Fedora, maybe SuSE and Adobe hammer out a deal so that SuSE pays Adobe themselves so they can include Photoshop in their free as in beer distro massively increasing their popularty and install base.
There are plenty of Linux vendors Adobe can talk to, and given the right offer I'm sure some of them would take up the deal. As long as that's the case I don't see any problems. If Debian somehow had a monopoly on Linux distributions, and no one could fork off to create a proprietary Debian (that's a couple of exceptionally large ifs) then there might be an issue because Adobe wouldn't have any options. That is never going to happen though (mostly because Debian will always be forkable - just look at all the commercial Debian based distros: Xandros, Linspire etc.) I don't think there are any issues here other those you make up.
Physicists know about the fundamental particles or the nature of space only through the mathematics that model the phenomena. Which is not to say that such English language renderings are useless, but they skillfully devise to distance themselves from what physicists actually do, as well as to reenforce readers' natural aversion to numbers and formulae.
Mathematics is just another language that has a certain felicity in explaining "structure". What you can say easily in mathematics may be considerably more complicated in another language. What do I mean by structure? Given a set you can apply structure to it - be that a topological or geometrical structure via concepts like open sets, or distances (metrics), or an algebraic or categorical structure via operations and mappings. Mathematics is all about describing and explaining that structure in a simple and compact way. The difficulty is that layer upon layer upon layer of abbreviation and compact notation has been built up so there's a lot of language to learn. At the same time the concepts it is describing are often exceedingly abstract - at times well outside anything you could experience - which makes it a difficult language to learn. In the end though, it is still just another language, and given its strengths a language well worth learning.
You can write haiku in languages other than japanese (and many people try) but without the syllabic structure and the tidy succinctness of the language itself it just isn't the same (though I have wondered about writing haiku in finnish). You can explain the deep structures of the universe in a language other than mathematics (and many people try) but without the many layers of abbreviation and the level of abstraction it will be an unbelievably long work that would take decades to read let alone write.
After OO2 is released, probably someone will fork it, replacing all the java, and call it FireOffice, then OpenOffice will adopt the changes.
I think, more importantly, any fork will instead be a word processor project. Oh, and a spreadsheet project. Possibly another project focussed on presentations. They might even bother to have a project for drawing and another for databases.
Firefox and Thunderbird are good - and more popular than the mozilla suite - because they are focussed on a specific task rather than bein a monolithic project trying to be everything at once. I would dearly love to see OpenOffice.org similarly broken up into more focussed projects.
Just to make sure I understand you: That means for every $1.00 earned on average $1.02 was spent? If so, that's horrible.
I don't have the 2002 figures to know if that's actually the case, but it isn't exactly out of the question. Certainly US savings rates have been abysmal. US household debt stood at $8,454,400,000,000 in 2002, here's a nice chart showing the acceleration into debt: http://www.globalpolicy.org/socecon/crisis/2003/ho usedebtchart.htm>
In 2004 household debt increased 4 times faster than the economy, and average credit card debt for households with at least 1 credit card increased 300%, to over $9000.
Given stats like that I can quite easily believe that personal savings actually dipped into the negatives one year.
You can find a Beckham jersey in pratically every country in the world. Especially in Asian countries.
If we're talking about South Asia and mentioning cricket (as the grandparent post did) then you'd be better of pointing to Sachin Tendulkar, Shoaib Ahktar, Sourav Ganguly, Muttiah Muralitharan and Shahid Afridi (with his stunning century against India a month or so ago). There's plenty of focus on these guys in India and Pakistan and Sri Lanka.
Let's put it another way: when the crincinfo servers seem to be running a bit slow (something I presume most people in the US don't generally experience) you can always use the Indian mirror - it has more bandwidth than you could possibly imagine: an attempt at Slashdotting wouldn't even make it blink. And for good reason - it gets truly unbeleivable amounts of traffic.
Heh. It's worse than that - the lack of personal savings is starting to cause serious economic issues: the current account deficit is blowing out and due to the lack of personal savings combined with the continuing massive budget deficits the falling US Dollar is failing to make a difference. The direction of the current account deficit has to change, and the causes of that change can either be cause a soft landing (an increase in personal savings combined with a balanced budget and a slowly falling dollar) or a hard one (a currency crisis, rampant inflation, and economic depression). The fact that people in the US are so disinclined toward personal savings, the soft landing scenario is very worrying.
Let's put it another way: a lot of the savings/investment is in the form of an expensive house - sure there's a large mortgage to pay off still, but housing prices are going up and up... Except the Fed is now steadily hiking interest rates to try and combat inflation. The real estate bubble could burst in another 6 months and all of a sudden a lot of people who were rich on paper are going to be a hell of a lot poorer on paper, but still have just as large a debt as they did before.
Either get working on personal savings, or get your money out of the US.
Jedidiah.
Re:Comments to come: blah, Perl hard to read/maint
on
Perl Medic
·
· Score: 3, Insightful
s/shitty perl code/shitty code in any language/;
You can inherit shit in any language.
That is true, but perl does provide easy ways to write sloppy or unmaintainable code. That's not necessarily a bad thing, because it is exactly that looseness that makes perl a very fast language to develop quick scripts in - there aren't anywhere near as many hoops to jump through and TIMTOWTDI makes it very easy to just write whatever you're thinking. For the niche of a language that lets you knock together something to do the job quickly perl is fine.
If you want anything maintainable you need to forget all of that and make use strict and use warnings mandatory, and enforce some coding style standards. That can be done, and if you do that it is easy enough to write maintainable perl - it's just that you can't expect the language to help you with it much: you have to enforce and police the standards yourself.
The guy down the hall starts to feel outmoded when me with my Linux machine and Betty with her Mac can interoperate together completely and can both interoperate with him, but his machine can't do the same.
Windows is about the only one that doesn't "play nice". You can give your research team (if you have one) some nice Sun workstations running Solaris, and run your servers on *BSD, and everything will still interoperate easily and play nice together. It isn't hard to have a heterogenous computing environment (different machines and OSs for different needs) and still have everything compatible and interoperating - as long as you don't add Windows to the mix.
A lot of the publishing work we do is in TeX because TeX is pretty much the defacto standard of professional publishing. MS has rudimentary at best support, whereas Linux and Mac make it easy as pie.
This was my problem when using Windows - not TeX specifically, but the general array of software that was useful or desireable for the work I was doing that was a pain on Windows. Mac, and Linux, and BSD, and Solaris all have a large amount of software that is just there "out of the box" and integrates easily into the system. Windows and Office offer the same integration, but that's all that's on offer on Windows.
I can't say I've ever entirely understood this attitude. I have used both GIMP and Photoshop a fair amount, and they both have their ups and downs. Certainly if you're doing any print work Photoshop is a given - ther's no comparison. Fromm there it really depends on what you're doing. In general I would say that Photoshop is probably ahead on most fronts, but the difference is far from night and day.
If I was back doing graphics professionally I'd definitely shell out for Photoshop - it has a number of small advantages and a little bit more polish, and who knows I may want to do some print work sometime to. As it stands I am doing other things for a living, but still have a need to do occasional image manipulation. If you think I'm paying for Photoshop in that situation you're dead wrong. There are really 2 markets here and each is big enough to support its own product. If you're a serious proffesional then Photoshop makes sense. If you're anything less than that the difference between GIMP and Photoshop is small enough that GIMP really is a good choice. Certainly I'd take GIMP over Photoshop Elements any day.
Having worked in the research department of a company that behaved like this I can relate. We had a small team in research, but they were very good people, and we were given a reasonably free hand to look into new areas and come up with new things. The problem was that the product people, and management, largely ignored (or equally often misintepreted) our ideas. It wasn't a communication issue, it was just that mamangement was extremely focused on the product that had and struggled to look outside it at all. I actually had a research project forcibly shut down because it was so much better than the current product. I had used some outside libraries to kickstart development, and it was a pure NIH decision.
In the end we had a huge blowup of a Dilbert cartoon posted in the office and directed people's queries there. It read something like:
PHB: This is Bob from the research department, he'll be presenting their latest findings. Bob: We at the research department have conducted a study of how our research gets used. Bob: We have found that all our research is either ignored, or misinterpreted by idiots, such as yourselves. Bob: Therefore, we have decided not to actually do any research anymore, we'll make stuff up. If you play along we'll make sure the comparative salary survey goes your way. Bob: Anyway, it's 3 o'clock, and that's quitting time in the research department. Wally (to Dilbert): I have a new hero.
At work, I saw 2000 BSOD on several servers when we applied an MS hotfix that conflicted with some sort of secret kernel patch they'd given us a few years previously for those same machines.
I saw 2k bluescreen one other time, when a workstation had a zip drive and the user installed drivers for it from 1997 or so.
For a while I had Win2k reproducibly bluescreening whenever I cared to. It was a laptop with a manufacturer install on Win2k, so we can presume drivers are not likely to be an issue, and the nature of it didn't imply any hardware issues.
In general Win2k was very solid for the time I was using it but I have had issues with it at times. I expect the problem I had got cleared up with the next service pack, but I wasn't using Win2k by then, so it wasn't of concern to me.
I have personally witnessed far more Win2k and XP Blue Screens than I have Linux or Solaris kernel panics (and I have seen all of the above at various times, and yes, hardware was the issue for at least some in each category).
Bring more stylized movement into the game, more theatrics and less of the old "run and hop" movement.
I think it would become boring quickly, if it's just eye candy and doesn't add to the gameplay. Counter-strike isn't much to look at but has good gameplay, and that is the reason it's still popular.
On the contrary, I think it could make all the difference. Some of the best parts of Doom3 were the animations and movements of the imps climbing walls and ceilings etc. Now imagine if creatures, characters, and so on, had some understanding of the physical environment and moved accordingly.
I don't mean ducking behind a pile of boxes, I mean using fluid animations to leap over piles of boxes instead of skirting around. I mean having the imps able to climb walls all the time. There's lots that could be done here to make character and monster movements more dynamic, varied and interesting, thus making the environment more immersing. All of this is, of course, very hard. Plenty of scope for improvement though.
Under fairer circumstances, who knows, IIS might have still won, but this rigged benchmark has nothing to offer us in deciding which server is faster.
I've reached the point where I completely ignore all the studies and benchmarks like this, from both sides. It is, quite simply, far too easy to set the constraints and metrics up so as to make sure you come out ahead. What's worse, it has become absolutely standard practice to do so. Studies have become completely useless because you can guarantee that they've been cooked one way or another.
Sadly, the actual format it's submitted in does matter, and not so much for the look.
I usually use paper myself, which means format is quite irrelevant, unless of course it is some format that is incapable of being printed.
Besides, unless you're applying as a graphics designer job or something like that, experience, knowledge and interview skills will matter a lot more than some fancy looking resume.
Actually I was applying for a job as a mathematician, so having a resume that was prepared by someone who obviously knew their way around TeX. Sure interview skills matter. Managing to get your resume noticed to get an interview also helps.
I actually find LaTeX resumés have the subtle advantage that they just look better. No, seriously. Everyone does their resumés in Word, and it isn't hard to spot Word documents, no matter how you mess with fonts. A LaTeX document just looks different - a little cleaner and sharper and more like professional typesetting.
Anything that can make your resumé stand out from the others in a good way is well worth pursuing.
MS Office is bloated, awkward and confusing. They need to make it *better* than MS Office. Do something innovative, instead of just copying.
I don't know how well Apple's iWork is selling (I heard not so well), but it's a hell of a lot nicer to use than Office because they looked at it from a different angle. It's missing some stuff, but Pages is a hell of a nice app for version 1.0.
It's a nice thought, but unfortunately there are different markets beign targetted here. I know it doesn't look like it: a word processor is a word processor right? The catch is that OpenOffice.org is trying to be the "all things to all people" office suite for corporate use. You don't get very far in that market unless you have [obscure feature foo] because apparently it is necessary for [small coprorate division bar] to do their job properly. By the time you've crammed in all the features demanded by all the different coporate grups with all their obscure individual needs there's not much else you cna have but a big bloated mess.
Pages takes the easy out and simply tries to be a nice, reasonably featureful word processor. By taking the step back and not trying to take over the "it needs to do everything" corporate market it can have a simple clean interface - just include the basic features. Abiword actually does a surprisingly good job of this for the open source crowd. It's still a little on the "crammed toolbar" side of things, but as competition with OpenOffice.org continues they may well find a simpler cleaner approach may well be to their benefit.
What it should have been is a War Hero who saved the Galaxy. THEN he got hungry with power.
Actually thinking about this part a little more, I think you are again on to something. One could easily develop a story a little akin to say Red Son which does a fine job of turning the all powerful hero into something of a villain in an elegant and believable way. Just add some material about Anakin's rage at not being able to control everything the way he feels he should/deserves to (perhaps the death of his wife could come in here somewhere) and thus falling to the dark side and you would have a much more compeeling story.
What should have happend was AOTC as Ep 1. Then Ep 2 can cover the Clone Wars, with ROTS still being Ep 3. That would have been a more effective story because it would have built Anakin up as a hero more, making his Fall greater.
This is actually rather insightful. If Episode 1 had featured Anakin being roughly Luke's age or a little younger then, for instance, the love story with whoever was going to be Luke's mother could have been introduced or strongly hinted at in Ep1, properly developed in an unhurried way in Ep2, then suitably tragically broken in Ep3.
The impression I had gotten from Ep4-6 was that Obi Wan had found himself out in the ass end of space, mostly out of touch with the Jedi council, and had encountered this great pilot with a lot of natural ability with the force. Unable to discuss the matter with the Jedi council Obi wan took it upon himself (still young and inexperienced at the time) to train Anakin, fucked it up, and fell right into the hands of the empire. That would have made some sense to me, and would have offered, as you say, a great chance to really develop Anakin as a heor, to really develop the love story over a couple of films, and to contrast the way in which Obi Wan teaches Luke with the way he teaches Anakin. Ah well, too late now.
Applauding the murder of abortion doctors is hardly mainstream american christian conventional wisdom. Have you ever talked to a christian?
And much of what the Taliban did was hardly mainstream Afghani muslim conventional wisdom. Have you even talked to a muslim?
The Taliban practiced a particularly severe and fundamentalist version of Islam that is in no way representative of Islam in general, nor even of Afghani muslim's in general. The Taliban are gone, many Afghani's celebrate that fact, but they are all still muslims, and they all still practice their beliefs - they just aren't the extremnist interpretations of the Taliban anymore.
The problem with the majority of Christians in this country is that they worship the Bible instead of God. Instead of asking themselves "What would a kind and compassionate God think of homosexuality (for example)" they point to the Bible and say "look- it says here that that is wrong!"
The problem is that they expect someone else to tell them what is right and what is wrong rather than thinking about the problem seriously themselves. The problem is that they rely on the crutches of inflexible unchanging antiquated religious morals to try and define their positions on moral problems that simply weren't relevant so long ago. The problem is that they expect to be taught, and they want to teach other people how to be taught.
I read a book on rational Zen once, and it had a very interesting explanation: the seemingly bizarre and self contradictory koans so often associated with zen stem from the fundamental contradiction that they are trying to teach people to not be taught. Oddly science and mathematics and a lot of philosophy has a similar belief, they just don't place it as being quite as fundamental. I think it is this belief that truly divides science from Christianity, and it is worth noting that it is not a divide between science and religion - there are religions that embrace such ideals.
Slashdot Mirror
Heh, interesting they chose Mathematica. Of course it was easy to port: it already ran on multiple processors (x86 included of course) and a bunch of OSes (OSX, Win, Linux). The changes were probably just a bunch of #define's.
And even then it sounds better than it is. Yes Mathematica is a very large complex application with vast amounts of code. Most of that code is written in Mathematica itself however, and requires no porting whatsoever if you can get the Mathematica interpreter up and running.
Think of it as like porting Emacs: you only really have to get the elisp engine running, and most everything else will then instantly fall into place. Yes that takes some work, but vastly less work than it might at first look like based on the apparent complexity of the application.
Jedidiah.
It doesn't shock me too much that it only took 2 hours to port Mathematica. I mean, the API for OS X on Intel is probably exactly the same as for OS X on PPC. Probably only very, very small parts (if any at all) of Mathematica are written in assembly code. You fix those parts and anything that relies on specific processor behavior then do a recompile.
The majority of Mathematica is written in Mathematica. Porting Mathematica over is probably akin to porting Emacs: you get the elisp going, and you're pretty much done.
Sure there is some code to port over, but remember that Mathematica runs on Windows, MacOS X, Linux, and Solaris, and that the majority of the code is in Mathematica and doesn't need to be changed... I'm a little surprised it took 2 hours.
Jedidiah.
Who among us doesn't have files saved in an old version of, say, Word, which can no longer be read correctly in a newer version of Word?
Not really. I do have a lot of rather old TeX documents from an older version of TeX, some are 10 or 15 years old. All of them produce results unerringly identical to the old version. I fully expect to be able to use the TeX of 10 or 15 years in the future to still produce exactly the same output from the source. I expect similar things could be said about nroff and the like, but I don't use those, so I can't say for sure.
The really interesting thing is that I now use pdftex on the documents to get PDF output instead of DVI and Postscript. I'm sure if a new display format becomes similarly popular in the future there'll be a version of TeX to compile directly to the new format as well.
I don't fully understand people locking away their documents the way they do...
Jedidiah.
I think this is pretty much the whole point: WebCore is a fork. There isn't much cooperation, or joint effort between WebCore and KHTML. Sure, some code manages to go both ways, but slowly but surely the two projects are diverging more and more. This isn't really a problem, the problem is expectations. The KDE people are pissed not so much at Apple, but rather at all the people that keep expecting changes in WebCore to come to KHTML. They are pissed with all the people that don't seem to understand that WebCore is a fork, and is now quite a separate project.
Go read some of the antiquated mailinglists for GNU/Emacs, I'm sure you'll find some equally pissed developers there complaining about people constantly asking when XEmacs changes will get merged into GNU/Emacs. That's all we're dealing with here.
Jedidiah.
Uh, yeah Bill, we've heard this promise before. I'm not holding my breath over any Microsoft promise that ends with "a thing of the past." The past keeps coming back to haunt you with Windows.
/editing security policies.
I think the thing to note is "without the users' knowledge" - I think we can still expect plenty of malware, viruses, and whatnot, they will simply be using social engineering attacks which, let's be frank, have proved remarkably effective. How many email viruses actually required user intervention to work? Building a system that is actually resistent to that is the real quest here.
The best solution to these sorts of problems is trusted computing and that comes in two flavours: the one provided by Microsoft and NGSCB, the one provided by TrustedSolaris, SELinux, TrustedBSD etc. The first locks down all the resources only giving access according to predefined rules: the general purpose computer becomes an appliance; it has a hardwired set of things that it can do. The second locks down resources only giving access according to an editable policy: if you really want to do something you can change the policy; you still have a general purpose computer.
The first flavour can, in theory, block the majority of social engineering attacks, but could become unusably constrained in so doing. The second flavour simply adds an extra layer to social engineering attacks: you have to convince the user to alter their policy as well as whatever steps were required for your malware. On the other hand, you can have a much tighter default policy.
Personally I like the second approach - I think SELinux looks excellent, and I am very keen to see OpenSolaris which includes TrustedSolaris. What the community needs to do is write/modify software to better respect the constraints that such systems allow, write good default policies, and create some good simple GUI tools for dealing with
Jedidiah.
Adobe will have much less chance of making money on the Linux market, because the Gimp is preinstalled on so many machines. This breaks the 3rd party market on Linux machines.
Not really no. It breaks the 3rd party market for Debian and other distros that want to be 100% FOSS, but other distros will happily bundle proprietary 3rd party apps. Linspire, for instance, bundles a payware DVD player.
If Adobe wanted to get into the Linux market they can go and chat to Redhat and SuSE and Sun and Linspire and set up a deal to have Photoshop for Linux included in the distro. How that deal pans out is unknown - maybe it only gets included in RHEL but not Fedora, maybe SuSE and Adobe hammer out a deal so that SuSE pays Adobe themselves so they can include Photoshop in their free as in beer distro massively increasing their popularty and install base.
There are plenty of Linux vendors Adobe can talk to, and given the right offer I'm sure some of them would take up the deal. As long as that's the case I don't see any problems. If Debian somehow had a monopoly on Linux distributions, and no one could fork off to create a proprietary Debian (that's a couple of exceptionally large ifs) then there might be an issue because Adobe wouldn't have any options. That is never going to happen though (mostly because Debian will always be forkable - just look at all the commercial Debian based distros: Xandros, Linspire etc.) I don't think there are any issues here other those you make up.
Jedidiah.
Physicists know about the fundamental particles or the nature of space only through the mathematics that model the phenomena. Which is not to say that such English language renderings are useless, but they skillfully devise to distance themselves from what physicists actually do, as well as to reenforce readers' natural aversion to numbers and formulae.
Mathematics is just another language that has a certain felicity in explaining "structure". What you can say easily in mathematics may be considerably more complicated in another language. What do I mean by structure? Given a set you can apply structure to it - be that a topological or geometrical structure via concepts like open sets, or distances (metrics), or an algebraic or categorical structure via operations and mappings. Mathematics is all about describing and explaining that structure in a simple and compact way. The difficulty is that layer upon layer upon layer of abbreviation and compact notation has been built up so there's a lot of language to learn. At the same time the concepts it is describing are often exceedingly abstract - at times well outside anything you could experience - which makes it a difficult language to learn. In the end though, it is still just another language, and given its strengths a language well worth learning.
You can write haiku in languages other than japanese (and many people try) but without the syllabic structure and the tidy succinctness of the language itself it just isn't the same (though I have wondered about writing haiku in finnish). You can explain the deep structures of the universe in a language other than mathematics (and many people try) but without the many layers of abbreviation and the level of abstraction it will be an unbelievably long work that would take decades to read let alone write.
Jedidiah.
After OO2 is released, probably someone will fork it, replacing all the java, and call it FireOffice, then OpenOffice will adopt the changes.
I think, more importantly, any fork will instead be a word processor project. Oh, and a spreadsheet project. Possibly another project focussed on presentations. They might even bother to have a project for drawing and another for databases.
Firefox and Thunderbird are good - and more popular than the mozilla suite - because they are focussed on a specific task rather than bein a monolithic project trying to be everything at once. I would dearly love to see OpenOffice.org similarly broken up into more focussed projects.
Jedidiah.
Just to make sure I understand you: That means for every $1.00 earned on average $1.02 was spent? If so, that's horrible.
o usedebtchart.htm>
I don't have the 2002 figures to know if that's actually the case, but it isn't exactly out of the question. Certainly US savings rates have been abysmal. US household debt stood at $8,454,400,000,000 in 2002, here's a nice chart showing the acceleration into debt: http://www.globalpolicy.org/socecon/crisis/2003/h
In 2004 household debt increased 4 times faster than the economy, and average credit card debt for households with at least 1 credit card increased 300%, to over $9000.
Given stats like that I can quite easily believe that personal savings actually dipped into the negatives one year.
Jedidiah.
You can find a Beckham jersey in pratically every country in the world. Especially in Asian countries.
If we're talking about South Asia and mentioning cricket (as the grandparent post did) then you'd be better of pointing to Sachin Tendulkar, Shoaib Ahktar, Sourav Ganguly, Muttiah Muralitharan and Shahid Afridi (with his stunning century against India a month or so ago). There's plenty of focus on these guys in India and Pakistan and Sri Lanka.
Let's put it another way: when the crincinfo servers seem to be running a bit slow (something I presume most people in the US don't generally experience) you can always use the Indian mirror - it has more bandwidth than you could possibly imagine: an attempt at Slashdotting wouldn't even make it blink. And for good reason - it gets truly unbeleivable amounts of traffic.
Jedidiah.
Heh. It's worse than that - the lack of personal savings is starting to cause serious economic issues: the current account deficit is blowing out and due to the lack of personal savings combined with the continuing massive budget deficits the falling US Dollar is failing to make a difference. The direction of the current account deficit has to change, and the causes of that change can either be cause a soft landing (an increase in personal savings combined with a balanced budget and a slowly falling dollar) or a hard one (a currency crisis, rampant inflation, and economic depression). The fact that people in the US are so disinclined toward personal savings, the soft landing scenario is very worrying.
Let's put it another way: a lot of the savings/investment is in the form of an expensive house - sure there's a large mortgage to pay off still, but housing prices are going up and up... Except the Fed is now steadily hiking interest rates to try and combat inflation. The real estate bubble could burst in another 6 months and all of a sudden a lot of people who were rich on paper are going to be a hell of a lot poorer on paper, but still have just as large a debt as they did before.
Either get working on personal savings, or get your money out of the US.
Jedidiah.
s/shitty perl code/shitty code in any language/;
You can inherit shit in any language.
That is true, but perl does provide easy ways to write sloppy or unmaintainable code. That's not necessarily a bad thing, because it is exactly that looseness that makes perl a very fast language to develop quick scripts in - there aren't anywhere near as many hoops to jump through and TIMTOWTDI makes it very easy to just write whatever you're thinking. For the niche of a language that lets you knock together something to do the job quickly perl is fine.
If you want anything maintainable you need to forget all of that and make use strict and use warnings mandatory, and enforce some coding style standards. That can be done, and if you do that it is easy enough to write maintainable perl - it's just that you can't expect the language to help you with it much: you have to enforce and police the standards yourself.
Jedidiah.
The guy down the hall starts to feel outmoded when me with my Linux machine and Betty with her Mac can interoperate together completely and can both interoperate with him, but his machine can't do the same.
Windows is about the only one that doesn't "play nice". You can give your research team (if you have one) some nice Sun workstations running Solaris, and run your servers on *BSD, and everything will still interoperate easily and play nice together. It isn't hard to have a heterogenous computing environment (different machines and OSs for different needs) and still have everything compatible and interoperating - as long as you don't add Windows to the mix.
A lot of the publishing work we do is in TeX because TeX is pretty much the defacto standard of professional publishing. MS has rudimentary at best support, whereas Linux and Mac make it easy as pie.
This was my problem when using Windows - not TeX specifically, but the general array of software that was useful or desireable for the work I was doing that was a pain on Windows. Mac, and Linux, and BSD, and Solaris all have a large amount of software that is just there "out of the box" and integrates easily into the system. Windows and Office offer the same integration, but that's all that's on offer on Windows.
Jedidiah.
I can't say I've ever entirely understood this attitude. I have used both GIMP and Photoshop a fair amount, and they both have their ups and downs. Certainly if you're doing any print work Photoshop is a given - ther's no comparison. Fromm there it really depends on what you're doing. In general I would say that Photoshop is probably ahead on most fronts, but the difference is far from night and day.
If I was back doing graphics professionally I'd definitely shell out for Photoshop - it has a number of small advantages and a little bit more polish, and who knows I may want to do some print work sometime to. As it stands I am doing other things for a living, but still have a need to do occasional image manipulation. If you think I'm paying for Photoshop in that situation you're dead wrong. There are really 2 markets here and each is big enough to support its own product. If you're a serious proffesional then Photoshop makes sense. If you're anything less than that the difference between GIMP and Photoshop is small enough that GIMP really is a good choice. Certainly I'd take GIMP over Photoshop Elements any day.
Jedidiah.
Work from Research is ignored.
Having worked in the research department of a company that behaved like this I can relate. We had a small team in research, but they were very good people, and we were given a reasonably free hand to look into new areas and come up with new things. The problem was that the product people, and management, largely ignored (or equally often misintepreted) our ideas. It wasn't a communication issue, it was just that mamangement was extremely focused on the product that had and struggled to look outside it at all. I actually had a research project forcibly shut down because it was so much better than the current product. I had used some outside libraries to kickstart development, and it was a pure NIH decision.
In the end we had a huge blowup of a Dilbert cartoon posted in the office and directed people's queries there. It read something like:
PHB: This is Bob from the research department, he'll be presenting their latest findings.
Bob: We at the research department have conducted a study of how our research gets used.
Bob: We have found that all our research is either ignored, or misinterpreted by idiots, such as yourselves.
Bob: Therefore, we have decided not to actually do any research anymore, we'll make stuff up. If you play along we'll make sure the comparative salary survey goes your way.
Bob: Anyway, it's 3 o'clock, and that's quitting time in the research department.
Wally (to Dilbert): I have a new hero.
Jedidiah.
At work, I saw 2000 BSOD on several servers when we applied an MS hotfix that conflicted with some sort of secret kernel patch they'd given us a few years previously for those same machines.
I saw 2k bluescreen one other time, when a workstation had a zip drive and the user installed drivers for it from 1997 or so.
For a while I had Win2k reproducibly bluescreening whenever I cared to. It was a laptop with a manufacturer install on Win2k, so we can presume drivers are not likely to be an issue, and the nature of it didn't imply any hardware issues.
In general Win2k was very solid for the time I was using it but I have had issues with it at times. I expect the problem I had got cleared up with the next service pack, but I wasn't using Win2k by then, so it wasn't of concern to me.
I have personally witnessed far more Win2k and XP Blue Screens than I have Linux or Solaris kernel panics (and I have seen all of the above at various times, and yes, hardware was the issue for at least some in each category).
Jedidiah.
Jedidiah.
On the contrary, I think it could make all the difference. Some of the best parts of Doom3 were the animations and movements of the imps climbing walls and ceilings etc. Now imagine if creatures, characters, and so on, had some understanding of the physical environment and moved accordingly.
I don't mean ducking behind a pile of boxes, I mean using fluid animations to leap over piles of boxes instead of skirting around. I mean having the imps able to climb walls all the time. There's lots that could be done here to make character and monster movements more dynamic, varied and interesting, thus making the environment more immersing. All of this is, of course, very hard. Plenty of scope for improvement though.
Jedidiah.
Under fairer circumstances, who knows, IIS might have still won, but this rigged benchmark has nothing to offer us in deciding which server is faster.
I've reached the point where I completely ignore all the studies and benchmarks like this, from both sides. It is, quite simply, far too easy to set the constraints and metrics up so as to make sure you come out ahead. What's worse, it has become absolutely standard practice to do so. Studies have become completely useless because you can guarantee that they've been cooked one way or another.
Jedidiah.
Sadly, the actual format it's submitted in does matter, and not so much for the look.
I usually use paper myself, which means format is quite irrelevant, unless of course it is some format that is incapable of being printed.
Besides, unless you're applying as a graphics designer job or something like that, experience, knowledge and interview skills will matter a lot more than some fancy looking resume.
Actually I was applying for a job as a mathematician, so having a resume that was prepared by someone who obviously knew their way around TeX. Sure interview skills matter. Managing to get your resume noticed to get an interview also helps.
Jedidiah.
I actually find LaTeX resumés have the subtle advantage that they just look better. No, seriously. Everyone does their resumés in Word, and it isn't hard to spot Word documents, no matter how you mess with fonts. A LaTeX document just looks different - a little cleaner and sharper and more like professional typesetting.
Anything that can make your resumé stand out from the others in a good way is well worth pursuing.
Jedidiah.
MS Office is bloated, awkward and confusing. They need to make it *better* than MS Office. Do something innovative, instead of just copying.
I don't know how well Apple's iWork is selling (I heard not so well), but it's a hell of a lot nicer to use than Office because they looked at it from a different angle. It's missing some stuff, but Pages is a hell of a nice app for version 1.0.
It's a nice thought, but unfortunately there are different markets beign targetted here. I know it doesn't look like it: a word processor is a word processor right? The catch is that OpenOffice.org is trying to be the "all things to all people" office suite for corporate use. You don't get very far in that market unless you have [obscure feature foo] because apparently it is necessary for [small coprorate division bar] to do their job properly. By the time you've crammed in all the features demanded by all the different coporate grups with all their obscure individual needs there's not much else you cna have but a big bloated mess.
Pages takes the easy out and simply tries to be a nice, reasonably featureful word processor. By taking the step back and not trying to take over the "it needs to do everything" corporate market it can have a simple clean interface - just include the basic features. Abiword actually does a surprisingly good job of this for the open source crowd. It's still a little on the "crammed toolbar" side of things, but as competition with OpenOffice.org continues they may well find a simpler cleaner approach may well be to their benefit.
Jedidiah
What it should have been is a War Hero who saved the Galaxy. THEN he got hungry with power.
Actually thinking about this part a little more, I think you are again on to something. One could easily develop a story a little akin to say Red Son which does a fine job of turning the all powerful hero into something of a villain in an elegant and believable way. Just add some material about Anakin's rage at not being able to control everything the way he feels he should/deserves to (perhaps the death of his wife could come in here somewhere) and thus falling to the dark side and you would have a much more compeeling story.
Jedidiah.
What should have happend was AOTC as Ep 1. Then Ep 2 can cover the Clone Wars, with ROTS still being Ep 3. That would have been a more effective story because it would have built Anakin up as a hero more, making his Fall greater.
This is actually rather insightful. If Episode 1 had featured Anakin being roughly Luke's age or a little younger then, for instance, the love story with whoever was going to be Luke's mother could have been introduced or strongly hinted at in Ep1, properly developed in an unhurried way in Ep2, then suitably tragically broken in Ep3.
The impression I had gotten from Ep4-6 was that Obi Wan had found himself out in the ass end of space, mostly out of touch with the Jedi council, and had encountered this great pilot with a lot of natural ability with the force. Unable to discuss the matter with the Jedi council Obi wan took it upon himself (still young and inexperienced at the time) to train Anakin, fucked it up, and fell right into the hands of the empire. That would have made some sense to me, and would have offered, as you say, a great chance to really develop Anakin as a heor, to really develop the love story over a couple of films, and to contrast the way in which Obi Wan teaches Luke with the way he teaches Anakin. Ah well, too late now.
Jedidiah.
Applauding the murder of abortion doctors is hardly mainstream american christian conventional wisdom. Have you ever talked to a christian?
And much of what the Taliban did was hardly mainstream Afghani muslim conventional wisdom. Have you even talked to a muslim?
The Taliban practiced a particularly severe and fundamentalist version of Islam that is in no way representative of Islam in general, nor even of Afghani muslim's in general. The Taliban are gone, many Afghani's celebrate that fact, but they are all still muslims, and they all still practice their beliefs - they just aren't the extremnist interpretations of the Taliban anymore.
Jedidiah.
The problem with the majority of Christians in this country is that they worship the Bible instead of God. Instead of asking themselves "What would a kind and compassionate God think of homosexuality (for example)" they point to the Bible and say "look- it says here that that is wrong!"
The problem is that they expect someone else to tell them what is right and what is wrong rather than thinking about the problem seriously themselves. The problem is that they rely on the crutches of inflexible unchanging antiquated religious morals to try and define their positions on moral problems that simply weren't relevant so long ago. The problem is that they expect to be taught, and they want to teach other people how to be taught.
I read a book on rational Zen once, and it had a very interesting explanation: the seemingly bizarre and self contradictory koans so often associated with zen stem from the fundamental contradiction that they are trying to teach people to not be taught. Oddly science and mathematics and a lot of philosophy has a similar belief, they just don't place it as being quite as fundamental. I think it is this belief that truly divides science from Christianity, and it is worth noting that it is not a divide between science and religion - there are religions that embrace such ideals.
Jedidiah.