I'm stupid enough. If your only metric for choosing a card is the current state of the art, then yeah maybe some of those late model cards look kind of shabby now, especially to gamers. But not me. They weren't crap then, and if they meet needs now they still aren't crap. Personally, I'm not a gamer, but I'd like to find a nice minimalist card that can get my Mac G4 to run Quartz Extreme graphics. I'm interested, but not to the point that I want to spend $200 or more on it, which as near as I can tell is what the minimum hardware costs at this point. If it fell to maybe a quarter of that price I'd be willing to go for it, but for the work I do -- where almost all of it is in a unix shell anyway -- it doesn't make sense to spend that kind of money on video hardware, even if it would make things prettier. But for $200, that's a big enough fraction of the way to a whole new computer that I personally just don't see the point.
So please, just because someone has different priorities than you do doesn't mean that they're stupid.
Just to pick nits, aside from Exchange server you just listed a bunch of non-Microsoft programs. Now granted, they all need Windows to run [will Wine do? I honestly don't know], but Quicken is already an alternative to MS Money, MS afaik doesn't offer media software like Cubase or Premier, Adobe themselves offer versions of Acrobat [dunno about more than reader, you may be way ahead of me here] for Win/Mac/Linux/Solaris, and Exchange, well, aside from Domino/Notes there is no major alternatives, and all the DIY approaches fall down for want of calendaring [but keep an eye on iCal, in terms of both the protocol/format and the Apple application -- that may be the best hope for an Exchange/Domino killer that I know of today].
It's not exactly ideal, but you might be able to get away with a Mac and VirtualPC. That'll get you native versions of most of the software you want, and you'll be able to run Quicken in emulation mode. If you need to run Exchange, well, I *guess* you could do that in emulation, but it would be more of a stretch. But anyway, for the other stuff you should be able to get by with OSX/VirtualPC.
'course, if you'd actually read the PHP presentation that the Yahoo people wrote, you'd see that they're making the same point as the poster you're objecting to. If you follow the link labelled "Why not JSP, Servlets, or J2EE?" [sorry, the annoying combination of frames & javascript keeps me from providing a more direct link] you'll see that their objections are the same ones that are made again here. In particular, Yahoo has been a FreeBSD shop for a long time now, and that drives much of the rest of the decision here. Quoting their slides:
Why not JSP, Servlets, or J2EE?
Pros
strongly typed
good performance (JIT), sandboxing
works w/lots of off-the-shelf software
But... you can't really use Java w/o threads
Threads support on FreeBSD is not great
If you read the slide on "Language Criteria", support for FreeBSD is one of the top points raised. The points about being a memory hog & non-portable are kind of a troll, yes (who cares if it's portable if it only has to run on your own servers?), but the thread thing is a deal breaker here.
Don't get me wrong, my pet language (Perl) wasn't chosen either, and I'd also quibble with the reasoning used there too (throwing "There's More Than One Way To Do It" back as a counterargument is kind of silly, and while "wasn't designed as web scripting language" is strictly speaking true it doesn't mean that it it's not a great web scripting language anyway [it is, in my opinion]).
But hey, get the facts, accept your pet language's flaws, and play to the strengths. And most of all, don't feed the trolls:-) -- photon317 was inflammatory in what he said, but he was on the right track and, most importantly, he nailed Java's achilles heel for this context. If the Yahoo people have done their homework & decided that PHP is "there yet" for "enterprise level" work -- and from the looks of it, they have done their homework -- then I'll accept their judgement even if I would personally go with something else. This is a big vote of confidence for PHP, and it wasn't given frivolously. Please try to accept that.
d) Try the tests while logged into the Mac in console mode, by typing
>console"
at the system login screen.
As others have noted, a dormant window manager shouldn't consume any processor time, but if you can disable it on both machines that's a more accurate comparison than trying to get an X11/KDE combination to perform similarly to Aqua. That itself would be a long, complex, and ultimately probably not very interesting comparison to run -- suffice to say that if you can make both window servers fall out of the picture the comparison should be more accurate.
Man, do all the big sites run on animals? What is the deal here? As high tech open source sites evolve over time, do they replace more and more of their systems with components designed by The Flintstones?
You really think that being able to cleanly translate from PDF to ASCII is going to be a step towards being able to modify a document? Surely I must be misunderstanding you, because obviously that isn't going to help at all after it throws away all the formatting, layout, images, etc.
This really isn't a platform issue -- download a copy of Cygwin & it can easily be used to install Ghostscript, rendering the two platforms isomorphic for these purposes. (And playing more into that point, Antiword makes it easy to convert DOCs to ASCII on *nix, so the conversion you want can go either way on either OS.) It's not that the tools aren't commonly used on the dominant platform, it's that the tools you're suggesting won't allow invisible tampering on *any* platform.
The point the original poster was getting at is that the PDF format is difficult to modify; it's much more tamper-proof than Word, and this implicit copy protection makes it a more desirable distribution format. The fact that this doesn't require people to shell out money for a copy of Word probably doesn't enter in the equation in the first place -- DOC can be modified, so DOC isn't suitable for distributing documents that aren't meant to be modified.
You can use Ghostscript to go from PDF to ASCII, or you can use Antiword to go from DOC to ASCII, but how are you going to get that ASCII document back into a form that is nearly identical to the original? I'd argue that you can't, because both conversations throw too much information away, and so both are going about the problem the wrong way.
An obvious forgery, there aren't nearly enough typos, grammatical errors, garbled logic, etc. It is well known that the real GWB write at a 2nd grade level, but this is solid 5th grade stuff. Nice try...:-)
What, so "press" is the Arabic word for a journalistic contact address now? What a cognate!!
You make a valid point, but English does seem to be the lingua franca of the interweb, even (apparently) among contries at the "axes of evil". The site seems to largely be in English, so the people running it presumably are English speakers as well. I can say, just from some of the foreign-born students I've known, that people that learn a technical subject in a particular language will tend to think in that language when practicing the craft, even if otherwise they speak something else. (For example, a Russian friend who studied aeronautical engineering as his father did, but couldn't discuss the subject with his dad because he only knew the English terms for everything & didn't know how to express the same concepts in his native language.)
So, like I say, I think your point is insightful, but at the same time I don't think it's unreasonable that the un/pw would have been English terms if the rest of the site was also English (as, from the little I poked around, it seems to be).
Actually, the comments made it look like you're meant to run light text on a dark background, which probably explains the luminence hack.
On a related note, I found out about a color version of the software, available (again with source etc) from this site. I'm sure this can be pushed in all sorts of ways (command line flags for color or "greyscale" or without color afjustment, command line hints for the desired image geometry, etc), but I don't know nearly enough C to feel comfortable with such a project (being a lowly Perl hacker & hardly an expert at that).
Other observations: [a] it can take any file that Quicktime can present, including video & still images (the thought that this program or something like it could attempt to render audio into an ascii movie is too terrifying & wonderful for me to grapple with yet), [b] you should be able to save the output it produces with/usr/bin/script, save it as a "plain" text file, and then play it back on any vt100ish terminal application, up to & including vt100 terminals if you have any laying around. This too is almost too terrible & wonderful for me to properly grapple with yet:-)
While you're upgrading your copy of Quicktime, you might also want to poke around Apple's developer's site & download their ASCIIMoviePlayer.
Yes, that's right -- in just 200 or so well documented lines of C, Apple shows you how to get your Mac's Quicktime libraries to output high quality video as....ascii. Eat your heart out, "Star Wars ascii guy" -- ever feel let down that you couldn't watch your home movie collection from work? Now all you need to do is open up a remote shell & play them in your Terminal/xterm/PuTTY window...
:-)
obReference: found out about this from MacOSXHints.com earlier this week, and have been amused by it ever since...
I won't bother with a point by point rebuttal of your trolling, especially since Dan has just countered all of your half-truths much better than I ever could. But one remark in particular stands out & bears a reply:
non-perl languages expected to provide additional support in the form of C code libraries for their opcodes.
If I remember the talks Dan has given to Boston.pm correctly, you've got this almost completely backwards. A more accurate reading is that Parrot is designed to act as kind of a superset of the target languages (Perl, Python, and Ruby) and that, in the handful of cases where an abstracted general feature of Parrot doesn't map perfectly back to the languages, any of those languages -- *including Perl* -- may have to adapt to Parrot's needs. In other words, Parrot has certain features because the Python & Ruby folks needed them, and these will require Perl to adapt if it wants to take advantage. (If we're lucky, Dan may clarify &/or add to this -- it's been a few months now and I forget most of the details of the presentation...).
So let's not have this kind of trolling, please. If you think Parrot is making such fundamental mistakes, help out or fork off your own project to prove yourself. Antagonistically spouting off half-correct technical gibberish on Slashdot isn't helping anyone.
You seem to have fallen down a rabbit hole. Did he say drive? Whoops, he meant train ride -- take the Acela and from what I hear it'll arrive in around two hours.
I mean yeah, I also thought the remark was funny, but it was obviously made in an offhand way as part of a much larger & more seriously meant point. Give the guy a break already, Slashdot has enough vitriol...:-)
Well, yes. Subjectively -- I'm no HI expert, just an interested amateur -- the Aqua guidelines seem much less polished & refined than the old Mac ones did a decade or more ago. Among the things that need to be cleaned up & made more coherent & cohesive are the various Mac vs. BSD artifacts: file paths, standard line endings, etc. My hope is that, as OSX & Aqua evolve, things will improve & get better standardized, but we'll see what happens.
But yeah, foreign interfaces are a plague that should almost always be avoided. Note how Photoshop & family are a dream on Macs, but IMO have never really fit in on PCs. Note how Microsoft adapted Office & Internet Explorer to the Mac from their original Windows versions, and for that matter Palm Desktop. Note how "cross platform" applications like Lotus Notes just looks, well, awful on every platform I've seen it on, and for that matter, most any GUI application written in Java/AWT or TCL/TK. Yuck!
Like I think I said in the earlier post, the only application I can think of where it brings the same interface to every platform and still works is Gvim -- but if you're the sort of person that's using Gvim then you probably aren't looking for aesthetics or goo menu layouts in the first place:)
So anyway, IMO the best thing the OpenOffice people can do is come up with a native Aqua/Cocoa interface to their software engine, much as the Chimera people have done with Gecko. The interface can & should draw from the Aqua guidelines if possible, but as Anarkhos says, getting the look but not the feel could be even worse than staying as an X11 app. IMO, anyway...
The principle that you're getting at is a sound one, and I don't object to it. That said though, I don't agree that the Aqua HIG document should be taken as the canonical reference for development on any platform. Rather, the Aqua HIG is a pretty good implementation of a more abstract general idea, but in other contexts different implementations of the idea can make more sense. For example, why should Windows developers try to adhere to the Mac user interface guidelines when Microsoft already publishers The Windows Interface Guidelines for Software Design: An Application Design Guide, which would be far more appropriate for Windows developers.
The thing is, user interface standards are, I think everyone would agree, generally a good thing. But different platforms have different ideas about what the standards should be. Any particular standard isn't necessarily better or worse than any of the others as long as it's consistent, logical, and easily learned. And as long as one standard is consistent & logical & so on, it's more of a problem to try to impose another systems consistent, logical framework. So for example the Aqua guidelines suggest how to arrange menus, what functionality should be in each one, how to arrange dialog windows, keystrokes to adhere to, etc. The Windows guidelines make different recommendations in each of these areas. The goal should be to adapt to the local system, so that users don't have to adapt to whatever platform the software was developed on.
The shambling train wreck here is Linux and X11, where the best you can hope for is some particular toolkits suggested but generally half thought out HIG standard. The best you can hope for is what Gnome or KDE offers, but still you as a user can't assume that all applications you use are going to adhere to one, the other, or even any standard. Everyone just makes up their own damn standard and the user has no choice but to wrestle them all down. Here, maybe it *would* make some sense to bring in ideas from the Aqua guidelines, or for that matter the Windows guidelines, the classic Mac guidelines, or hell anything else -- just pick *anything* and implement it *consistently*. But of course this has never happened and at this point I don't expect it to ever come together, short of a miracle in say RedHat's effort to merge KDE & Gnome. More power to 'em I say.
Anyway, I think what you really want is for someone to approach this as a true & complete discipline, just as programming & QA & administration are all disciplines. We need system designers that understand general UI theory (including general design principles, user testing & feedback schemes, etc) as well as specific implementations of the general theory as seen in e.g. Aqua, Windows, web design [Jakob Nielsen type stuff], etc. But in the end this all just has to be source material, and short of adopting someone else's standards full out -- that'll never happen -- in the end a cohesive Linux/X11 UI standard needs to emerge. Gnome & KDE & similar projects will play into this of course, but even those aren't fleshed out enough and the pointless rift between the two projects doesn't help things anyway. As long as there continues not to be a well thought out Linux/X11 HIG document that is widely referred to & implemented, using Linux will continue to be a painful experience for average [read: non-geek] users.
Xserve's price/performance ratio is pretty good, as is its absolute performance, in my oppinion.
Well thank you for your opinion, but my company did some comparison shopping & ended up deciding that for the services we'd want to run -- http, smtp, dns, cvs, etc -- we could get Dell or Sun blade machines for significantly less money. Most of us like Macs a lot, but for a server we're not trying to do anything that would run better on an OSX server than it would on some flavor of Linux or Solaris, so we just haven't been able to justify the expense to the finance people.
This is not just from looking at screen shots. It's from comparing the specs of the Xserve and similarly powered &/or priced alternatives, and it seems like consistently you can either get the same amount of power for less money or you can get a more powerful machine for the same money, your pick. The Xserve might be sexier than the alternatives but then hey who are you trying to impress, other sysadmins? It's not like your SO is likely to care...:-)
It's not so much a matter of trusting Linux "more" than BSD to serve a database -- I'm pretty much agnostic there. The bigger question is whether there is a lightweight version of the software that would run happily on non server level machines. I'd have no problem running something on the scale of MySQL, PostgreSQL, FileMaker, or Access on something as powerful as the average Mac or Intel desktop machine; I would be more reluctant to run something on the magnitide of Oracle on the same hardware, and my impression, just to get back to the original point, was that DB2 is similarly heavyweight software. I'd be happy to be corrected/enlightened here.
Your request cannot be completed. The following error message was returned:
The survey you requested: "4551/DB2forMacOS.htm" has been deactivated.
For questions regarding the survey you were attempting to complete, please contact the owner of the survey directly.
If you feel that you are receiving this message as a result of a system error, please send us an email.
Anyway, I've never worked with DB2 before so yes I'm talking out of my ass here, but I picture it as a high quality enterprise server database meant to be run on big IBM mainframes, big Sun servers, medium size Dell or HP Linux machines, etc. I also have the impression that Apple's one offering this direction -- Xserve -- is a nice but overpriced & underuseful machine that isn't going to be replacing the competition any time soon. For better or worse, I picture OSX as a client OS, not a server one, though perhaps that will change over time. [NB that I'm typing this from an OSX box.]
So, all that said, does anyone run DB2 on client workstation hardware? I can maybe picture developers working with an intra-office instance of the server, but really I thought it wanted something more substantial than the average PC or Mac in order to run happily. Am I wrong? Would any of you have a use for client / lightweight DB2?
Can the Clie's finally sync with a Mac? If so, can they take advantage of iSync, the system address book, iTunes, iPhoto, iYadda iYadda iYadda? The Sony's are cool devices but if I can't plug them into my computer then it's not going to do me much good.
Can the Clie do Bluetooth, 802.11b, etc? Where can I find more about the "wireless networking"?
Where can I find more about PalmOS 5 in general and as Sony has used it? Does this version start to bring in BeOS technology yet? If so what?
These things look cool but I want to see specs before I get my hopes up...
They're getting bigger because the market is getting big enough to support that kind of diversity. There are still small, utilitarian devices out there -- I'd argue that this is where Palm began and, for the most part, has remained ever since. They've done a good job in that niche and that's fine -- I still use my Pam V every day and love it.
But there's also a place for more capable devices, as the PocketPC market illustrates, and in order to differentiate themselves from Palm & Handspring, it makes sense for Sony to offer this kind of device in addition to their simpler models. If all Sony did was offer a faithful replica of the Palm -- including Palm's software -- then why would anyone bother buying it? They have no choice but to try to stand out from the pack.
Personally, I think it's great. You're right -- I don't see the high end Clie's as [uber-] PDAs, I see them as mini laptops that you can fit in your pocket. I see these devices as being replacements for all the little gizmos I've been thinking about buying -- digital camera, portable mp3 player, voice recorder, and yes an updated pda. It's all of these, and that's great to me. You can accuse it of the old "jack of all trades, master of none" line of thought, and to an extent that's true -- there are better "pure" devices in each area where these machines offer functionality. But hell I'm willing to take a slightly less fancy version in order to get all that functionality into one sleek little machine. Hell yeah.
Have all of your workers wrestle out their own/etc/printcap entries and your office will go paperless remarkably quickly. Possible obstacles:
Some poor bastard will figure out how to point a print document at the printer, but will never -- this you can count on -- never ever figure out how to make it look even remotely correct. For a time this frustrated sap will burn through reams of paper before giving up in disgust
Some people will try to print from Mac or Windows computers. Do not allow this! It is far too easy to allow some unwitting boob get up & running with normal printing on these platforms, at which your noble ecological pipe dream will end up flushed like yesterday's half digested tofu burger.
Otherwise, really, that's about all their is to it -- normal Linux / Unix LP print services. Switch to that and you'll never have to replace your toner cartridges again!
Honestly, I'd been using news.google.com as my main headlines page for several months now, and was surprised to find, when I woke up & checked my regular morning pages the other day, that the layout had been made much more complicated (because, of course, the site had gone public).
The previous layout was a whole lot simpler, just a simple list of categories, top stories within each, and four or five links to that story from different sources. One nice touch was that the link for each story was the headline used for it, which was nice because you could tell at a glance who was just repeating a wire feed and who really had something worthwhile -- and sometimes you could get nicely contrasting stories (like, say, the same event in Kashmir as described by both Indian & Pakistani news sources). The new, more complex & busy layout doesn't allow them to do this anymore, which IMO is a change for the worse.
AS for the new layout, I dunno. It has much higher information density, which the Edward Tufte fan in me thinks is a very good thing. But it's a very busy layout, and so a bit overwhelming to me. I'm finding that I haven't spent as much time on the new version as I was before on the old one, and I'm not checking it as often either -- maybe just a cursory glance once or twice a day, as opposed to a more careful skim several times a day before. Compared to the sparse layouts that Google ordinarily uses, a design this heavy feels very jarring to me, where on another site I probably wouldn't care. Hopefully I'll get over this.
Here's an interesting angle though, from the article the original submitter noted:
Google News already has made arrangements with some leading news sites that use registration schemes -- such as The New York Times. Google News users who click on links to NYTimes.com articles at Google News go directly to the article -- there's no intervening registration screen -- even if they're not already registered at NYTimes.com. This works, explains product manager Mayer, because the site allows Google's spiders to crawl its content and include links in the Google service. When a non-registered user hits a NYTimes.com page, the site will recognize that it's a referral from Google News and serve up the content -- delaying the registration requirement for one page. When the Google News user tries to go elsewhere on NYTimes.com, then the registration system kicks in. If the user is already registered, then NYTimes.com reads the user's NYT cookie and doesn't ask for registration information.
Why can't Slashdot come up with such an arrangment? The NY TImes is one of the best news sources on the 'net, and I'm sure their staff has to have at least some Slashdot fans. The constant whining disclaimers about having to register -- and the even more bizarre constant opposition to the very idea -- could all be short-circuited if the two sites could enter into a similar arrangement. Why has this never happened? Lack of imagination, or is one side or the other just uninterested? Whatever the obstacle has been, I'd be happy if we could just get over it and set up some kind of arrangement.
Sorry if this is offtopic, but I've just upgraded my 2+ year old Mac G4 to OSX 10.2, and I really haven't seen any big performance gain by doing so. From what I've read, if your hardware can do Quartz Extreme, you should be able to get much better performance, but it seems like my graphics card is too old for it, and so Jaguar hasn't been able to perform significantly better than 10.1.
So, any suggestions about what the best, cheap upgrade car for a two year old Mac would be? It's not worth it to me to shell out $200 or more for the top of the line hardware -- I don't play video games or anything like that -- but if a video card in the say $50 to $75 range would give a noticeable boost then it might be worthwhile.
Does anyone know what the minimum video hardware is to get QE running and how much it would cost to get that hardware costs these days? On the same lines, given similar hardware, have people seen better gains by upgrading graphics hardware or adding more ram? For the money I'm willing to spend right now, I could throw in half a gig of ram, but I've heard that upgrading the video card could be almost as much of a performance boost. It would be nice to get a few more opinions on which upgrade path makes more sense...
Slashdot Mirror
So please, just because someone has different priorities than you do doesn't mean that they're stupid.
It's not exactly ideal, but you might be able to get away with a Mac and VirtualPC. That'll get you native versions of most of the software you want, and you'll be able to run Quicken in emulation mode. If you need to run Exchange, well, I *guess* you could do that in emulation, but it would be more of a stretch. But anyway, for the other stuff you should be able to get by with OSX/VirtualPC.
If you read the slide on "Language Criteria", support for FreeBSD is one of the top points raised. The points about being a memory hog & non-portable are kind of a troll, yes (who cares if it's portable if it only has to run on your own servers?), but the thread thing is a deal breaker here.
Don't get me wrong, my pet language (Perl) wasn't chosen either, and I'd also quibble with the reasoning used there too (throwing "There's More Than One Way To Do It" back as a counterargument is kind of silly, and while "wasn't designed as web scripting language" is strictly speaking true it doesn't mean that it it's not a great web scripting language anyway [it is, in my opinion]).
But hey, get the facts, accept your pet language's flaws, and play to the strengths. And most of all, don't feed the trolls :-) -- photon317 was inflammatory in what he said, but he was on the right track and, most importantly, he nailed Java's achilles heel for this context. If the Yahoo people have done their homework & decided that PHP is "there yet" for "enterprise level" work -- and from the looks of it, they have done their homework -- then I'll accept their judgement even if I would personally go with something else. This is a big vote of confidence for PHP, and it wasn't given frivolously. Please try to accept that.
Err, no quote on console there, damn typos...
d) Try the tests while logged into the Mac in console mode, by typing
at the system login screen.As others have noted, a dormant window manager shouldn't consume any processor time, but if you can disable it on both machines that's a more accurate comparison than trying to get an X11/KDE combination to perform similarly to Aqua. That itself would be a long, complex, and ultimately probably not very interesting comparison to run -- suffice to say that if you can make both window servers fall out of the picture the comparison should be more accurate.
:-)
This really isn't a platform issue -- download a copy of Cygwin & it can easily be used to install Ghostscript, rendering the two platforms isomorphic for these purposes. (And playing more into that point, Antiword makes it easy to convert DOCs to ASCII on *nix, so the conversion you want can go either way on either OS.) It's not that the tools aren't commonly used on the dominant platform, it's that the tools you're suggesting won't allow invisible tampering on *any* platform.
The point the original poster was getting at is that the PDF format is difficult to modify; it's much more tamper-proof than Word, and this implicit copy protection makes it a more desirable distribution format. The fact that this doesn't require people to shell out money for a copy of Word probably doesn't enter in the equation in the first place -- DOC can be modified, so DOC isn't suitable for distributing documents that aren't meant to be modified.
You can use Ghostscript to go from PDF to ASCII, or you can use Antiword to go from DOC to ASCII, but how are you going to get that ASCII document back into a form that is nearly identical to the original? I'd argue that you can't, because both conversations throw too much information away, and so both are going about the problem the wrong way.
Err, "writes".
And it's also well known that people against the Bush regime have their own well documented problems with getting straight A's in grammar too :-)
An obvious forgery, there aren't nearly enough typos, grammatical errors, garbled logic, etc. It is well known that the real GWB write at a 2nd grade level, but this is solid 5th grade stuff. Nice try... :-)
You make a valid point, but English does seem to be the lingua franca of the interweb, even (apparently) among contries at the "axes of evil". The site seems to largely be in English, so the people running it presumably are English speakers as well. I can say, just from some of the foreign-born students I've known, that people that learn a technical subject in a particular language will tend to think in that language when practicing the craft, even if otherwise they speak something else. (For example, a Russian friend who studied aeronautical engineering as his father did, but couldn't discuss the subject with his dad because he only knew the English terms for everything & didn't know how to express the same concepts in his native language.)
So, like I say, I think your point is insightful, but at the same time I don't think it's unreasonable that the un/pw would have been English terms if the rest of the site was also English (as, from the little I poked around, it seems to be).
On a related note, I found out about a color version of the software, available (again with source etc) from this site. I'm sure this can be pushed in all sorts of ways (command line flags for color or "greyscale" or without color afjustment, command line hints for the desired image geometry, etc), but I don't know nearly enough C to feel comfortable with such a project (being a lowly Perl hacker & hardly an expert at that).
Other observations: [a] it can take any file that Quicktime can present, including video & still images (the thought that this program or something like it could attempt to render audio into an ascii movie is too terrifying & wonderful for me to grapple with yet), [b] you should be able to save the output it produces with /usr/bin/script, save it as a "plain" text file, and then play it back on any vt100ish terminal application, up to & including vt100 terminals if you have any laying around. This too is almost too terrible & wonderful for me to properly grapple with yet :-)
Yes, that's right -- in just 200 or so well documented lines of C, Apple shows you how to get your Mac's Quicktime libraries to output high quality video as ....ascii. Eat your heart out, "Star Wars ascii guy" -- ever feel let down that you couldn't watch your home movie collection from work? Now all you need to do is open up a remote shell & play them in your Terminal/xterm/PuTTY window...
:-)
obReference: found out about this from MacOSXHints.com earlier this week, and have been amused by it ever since...
If I remember the talks Dan has given to Boston.pm correctly, you've got this almost completely backwards. A more accurate reading is that Parrot is designed to act as kind of a superset of the target languages (Perl, Python, and Ruby) and that, in the handful of cases where an abstracted general feature of Parrot doesn't map perfectly back to the languages, any of those languages -- *including Perl* -- may have to adapt to Parrot's needs. In other words, Parrot has certain features because the Python & Ruby folks needed them, and these will require Perl to adapt if it wants to take advantage. (If we're lucky, Dan may clarify &/or add to this -- it's been a few months now and I forget most of the details of the presentation...).
So let's not have this kind of trolling, please. If you think Parrot is making such fundamental mistakes, help out or fork off your own project to prove yourself. Antagonistically spouting off half-correct technical gibberish on Slashdot isn't helping anyone.
I mean yeah, I also thought the remark was funny, but it was obviously made in an offhand way as part of a much larger & more seriously meant point. Give the guy a break already, Slashdot has enough vitriol... :-)
it
was
a
joke
Like, chill out already... :-)
But yeah, foreign interfaces are a plague that should almost always be avoided. Note how Photoshop & family are a dream on Macs, but IMO have never really fit in on PCs. Note how Microsoft adapted Office & Internet Explorer to the Mac from their original Windows versions, and for that matter Palm Desktop. Note how "cross platform" applications like Lotus Notes just looks, well, awful on every platform I've seen it on, and for that matter, most any GUI application written in Java/AWT or TCL/TK. Yuck!
Like I think I said in the earlier post, the only application I can think of where it brings the same interface to every platform and still works is Gvim -- but if you're the sort of person that's using Gvim then you probably aren't looking for aesthetics or goo menu layouts in the first place :)
So anyway, IMO the best thing the OpenOffice people can do is come up with a native Aqua/Cocoa interface to their software engine, much as the Chimera people have done with Gecko. The interface can & should draw from the Aqua guidelines if possible, but as Anarkhos says, getting the look but not the feel could be even worse than staying as an X11 app. IMO, anyway...
The thing is, user interface standards are, I think everyone would agree, generally a good thing. But different platforms have different ideas about what the standards should be. Any particular standard isn't necessarily better or worse than any of the others as long as it's consistent, logical, and easily learned. And as long as one standard is consistent & logical & so on, it's more of a problem to try to impose another systems consistent, logical framework. So for example the Aqua guidelines suggest how to arrange menus, what functionality should be in each one, how to arrange dialog windows, keystrokes to adhere to, etc. The Windows guidelines make different recommendations in each of these areas. The goal should be to adapt to the local system, so that users don't have to adapt to whatever platform the software was developed on.
The shambling train wreck here is Linux and X11, where the best you can hope for is some particular toolkits suggested but generally half thought out HIG standard. The best you can hope for is what Gnome or KDE offers, but still you as a user can't assume that all applications you use are going to adhere to one, the other, or even any standard. Everyone just makes up their own damn standard and the user has no choice but to wrestle them all down. Here, maybe it *would* make some sense to bring in ideas from the Aqua guidelines, or for that matter the Windows guidelines, the classic Mac guidelines, or hell anything else -- just pick *anything* and implement it *consistently*. But of course this has never happened and at this point I don't expect it to ever come together, short of a miracle in say RedHat's effort to merge KDE & Gnome. More power to 'em I say.
Anyway, I think what you really want is for someone to approach this as a true & complete discipline, just as programming & QA & administration are all disciplines. We need system designers that understand general UI theory (including general design principles, user testing & feedback schemes, etc) as well as specific implementations of the general theory as seen in e.g. Aqua, Windows, web design [Jakob Nielsen type stuff], etc. But in the end this all just has to be source material, and short of adopting someone else's standards full out -- that'll never happen -- in the end a cohesive Linux/X11 UI standard needs to emerge. Gnome & KDE & similar projects will play into this of course, but even those aren't fleshed out enough and the pointless rift between the two projects doesn't help things anyway. As long as there continues not to be a well thought out Linux/X11 HIG document that is widely referred to & implemented, using Linux will continue to be a painful experience for average [read: non-geek] users.
But then we all know that already, don't we?
This is not just from looking at screen shots. It's from comparing the specs of the Xserve and similarly powered &/or priced alternatives, and it seems like consistently you can either get the same amount of power for less money or you can get a more powerful machine for the same money, your pick. The Xserve might be sexier than the alternatives but then hey who are you trying to impress, other sysadmins? It's not like your SO is likely to care... :-)
It's not so much a matter of trusting Linux "more" than BSD to serve a database -- I'm pretty much agnostic there. The bigger question is whether there is a lightweight version of the software that would run happily on non server level machines. I'd have no problem running something on the scale of MySQL, PostgreSQL, FileMaker, or Access on something as powerful as the average Mac or Intel desktop machine; I would be more reluctant to run something on the magnitide of Oracle on the same hardware, and my impression, just to get back to the original point, was that DB2 is similarly heavyweight software. I'd be happy to be corrected/enlightened here.
Anyway, I've never worked with DB2 before so yes I'm talking out of my ass here, but I picture it as a high quality enterprise server database meant to be run on big IBM mainframes, big Sun servers, medium size Dell or HP Linux machines, etc. I also have the impression that Apple's one offering this direction -- Xserve -- is a nice but overpriced & underuseful machine that isn't going to be replacing the competition any time soon. For better or worse, I picture OSX as a client OS, not a server one, though perhaps that will change over time. [NB that I'm typing this from an OSX box.]
So, all that said, does anyone run DB2 on client workstation hardware? I can maybe picture developers working with an intra-office instance of the server, but really I thought it wanted something more substantial than the average PC or Mac in order to run happily. Am I wrong? Would any of you have a use for client / lightweight DB2?
These things look cool but I want to see specs before I get my hopes up...
But there's also a place for more capable devices, as the PocketPC market illustrates, and in order to differentiate themselves from Palm & Handspring, it makes sense for Sony to offer this kind of device in addition to their simpler models. If all Sony did was offer a faithful replica of the Palm -- including Palm's software -- then why would anyone bother buying it? They have no choice but to try to stand out from the pack.
Personally, I think it's great. You're right -- I don't see the high end Clie's as [uber-] PDAs, I see them as mini laptops that you can fit in your pocket. I see these devices as being replacements for all the little gizmos I've been thinking about buying -- digital camera, portable mp3 player, voice recorder, and yes an updated pda. It's all of these, and that's great to me. You can accuse it of the old "jack of all trades, master of none" line of thought, and to an extent that's true -- there are better "pure" devices in each area where these machines offer functionality. But hell I'm willing to take a slightly less fancy version in order to get all that functionality into one sleek little machine. Hell yeah.
I know what I want for Christmas... :-)
But that's the beauty of the old lp system -- force people to switch to that and they'll never want to print anything ever again :-)
Otherwise, really, that's about all their is to it -- normal Linux / Unix LP print services. Switch to that and you'll never have to replace your toner cartridges again!
:-)
The previous layout was a whole lot simpler, just a simple list of categories, top stories within each, and four or five links to that story from different sources. One nice touch was that the link for each story was the headline used for it, which was nice because you could tell at a glance who was just repeating a wire feed and who really had something worthwhile -- and sometimes you could get nicely contrasting stories (like, say, the same event in Kashmir as described by both Indian & Pakistani news sources). The new, more complex & busy layout doesn't allow them to do this anymore, which IMO is a change for the worse.
AS for the new layout, I dunno. It has much higher information density, which the Edward Tufte fan in me thinks is a very good thing. But it's a very busy layout, and so a bit overwhelming to me. I'm finding that I haven't spent as much time on the new version as I was before on the old one, and I'm not checking it as often either -- maybe just a cursory glance once or twice a day, as opposed to a more careful skim several times a day before. Compared to the sparse layouts that Google ordinarily uses, a design this heavy feels very jarring to me, where on another site I probably wouldn't care. Hopefully I'll get over this.
Here's an interesting angle though, from the article the original submitter noted:
Why can't Slashdot come up with such an arrangment? The NY TImes is one of the best news sources on the 'net, and I'm sure their staff has to have at least some Slashdot fans. The constant whining disclaimers about having to register -- and the even more bizarre constant opposition to the very idea -- could all be short-circuited if the two sites could enter into a similar arrangement. Why has this never happened? Lack of imagination, or is one side or the other just uninterested? Whatever the obstacle has been, I'd be happy if we could just get over it and set up some kind of arrangement.
So, any suggestions about what the best, cheap upgrade car for a two year old Mac would be? It's not worth it to me to shell out $200 or more for the top of the line hardware -- I don't play video games or anything like that -- but if a video card in the say $50 to $75 range would give a noticeable boost then it might be worthwhile.
Does anyone know what the minimum video hardware is to get QE running and how much it would cost to get that hardware costs these days? On the same lines, given similar hardware, have people seen better gains by upgrading graphics hardware or adding more ram? For the money I'm willing to spend right now, I could throw in half a gig of ram, but I've heard that upgrading the video card could be almost as much of a performance boost. It would be nice to get a few more opinions on which upgrade path makes more sense...
Thanks!