Don't assume that I was arguing in favor of things being different than they are as such. I am not making decisions, merely presenting opinions and observations (that are more valid than you imply here).
I never said NASA was disfunctional in any way, though being a bureacracy (strike one) made up of humans (strike two) doesn't exactly inspire an high degree of confidence. But I feel that way with most human bureaucracies, so I learn to not worry about it. Certainly the (almost) perfect Cassini-Huygens work so far has made up greatly for many of the human errors of the Hubble first-try and Mars Probes, and the great unknown error in the Galileo probe's antenna.
I was not recommending that NASA become privatized, only that they private companies think ahead a little better and PLAN so that when things have to be replaced, the replacements are ready. I was not specifically saying that Hubble needs to be replaced ASAP (though I certainly would like it to, and I'm *hardly* alone on that, given what I've gleaned from friends who work in NASA). I'm just saying that they had nothing on the table for that eventuality, just as they currently have little on the table for the eventuality of eliminating the shuttle fleet.
That was all my comparison was meant to make, and I said nothing as such about desiring to privatize NASA and my follow-up post did mention the "profit" nature and how long-term research doesn't lead to profit in a recognizable sense.
Most companies, when they see a success, *immediately* begin plans for the next version. Software, entertainment, "inventions". They get a tangible benefit from something, see places where it might be better, and get started on building the next while another group continues to get what they can supporting the old.
NASA has done this before, but there are times when they act like they've never heard of this concept, and Hubble's fate to me is one of those times. Hubble dies in 2008, and the best chance we've been told for a replacement is 2020 (last I heard). THAT is lousy planning for a program most researchers consider a great success (once they got past the initial technical goof).
Granted, the crappy budget the Feds give them doesn't exactly help, being the result of the fact that Washington is just as short-sighted as Wall Street is, as demonstrated by the DARPA decision posted elsewhere on this site.
Planned obsolescense is an entirely different thing. It's not about making a better product, it's about making crappy products that break so they can sell you a replacement that doesn't do anything better than the old one.
yes, but it came about under the realization that even building for "the long term" only puts you out of business (you run out of customers as everybody who wants one has one), or people buy a new one anyways while their old one still works, just for aesthetic reasons.
No matter how it is, the fact that p.o. is practiced is a result of companies realizing, regardless of particular profit motives, that you have to have a new model ready before the old one is really obsolete, to keep the cash flow coming.
NASA should look at their stuff the same way -- only instead of cash, think "information". Hubble is going to stop sending data, and there's nothing going up there to send the same type of data back. The shuttles are going to be decomissioned, and there is nothing in the production stages that will be ready to take their place when it happens. The information stops, because they didn't plan enough on what to do when the suppliers of that information that seemed briefly so unlimited stopped functioning permanently.
I agree that *nobody* treats long-term research with any respect anymore, thanks to Wall Street's permanent emphasis on "quarterly results" at the expense of a company's real future. The final result of all this seems to be a world created *entirely* by amateurs.
acknowledging, of course, that as a commercial enterprise, NASA would have to find some way (besides selling satellite launch services) to actually turn a profit...and the curse is that long-term research isn't *meant* to turn a profit "now".
yeah. the *real* problem with NASA is they see these end-of-usefulness deadlines as being N years off, and forget that any replacement technology needs N+M years to develop, but because N years seems so far away, they don't start on the replacement...
so here we are with a shuttle fleet on the virge of permanent decommision (and 2 lost already) and a hubble satellite, and no plans for an actual replacement because N years hit NASA a LOT sooner than they realized...
Its the problem of NASA not being a commercial enterprise. The IT world, the auto world, the airplane world, the appliance world, the electronics world have all lived under the spectre of Plan Obsolescense for decades. We *know* that we have to have the replacement for something read *before* the current model goes out of usefulness.
NASA has never lived under that problem before, so as an institution, it simply didn't know how to react.
yeah, it really comes down to question of "what would you do with it". its all good and well that everybody likes feature "X" that comes from the integrated suite that currently the separated apps can't give (doesn't mean they won't get some form of triggering the other apps as a result of this announcement), but what more do you want besides a browser that stays the way you already know?
Mozilla SeaMonkey drove producing a set of capabilities cloning the original Netscape base it derived from (in a very detached way, of course), with the idea that Netscape/AOL (and others) could take the baseline and produce *standards-compliant* browsers on top of those capabilities.
3 things happened, and one thing *didn't* happen.
1) Netscape 6.x and 7.x were successfully rolled out based on the Mozilla baseline (only now 8.x has already started coming from the Firefox base, and the "Communicator" concept is gone)
2) AOL decided, in spite of their investment, to give up the idea of actually doing an AOL browser based on Mozilla in favor of playing marketting games with Microsoft by supporting IE instead in exchange for not being blacklisted off of Microsoft's illegal (but still practiced) OEM deals.
3) Firefox came out and had a marketting push unlike any other open source project around, including Linux distros themselves. And to top it off, the damn thing actually works (those few slashdotter complaints in this thread notwithstanding).
the thing that didn't happen: other ISPs didn't build their own browsers on Mozilla tech. It used to be in the 90s that all the ISPs followed AOL and Compuserve's lead in taking an existing browser (usually IE) and specializing it to become part of their service. Mozilla was setting up its code base specifically for that purpose -- we provide standards-compliance and ease-of-skinning; you skin it to meet your customer's needs. Only by the time Mozilla's codebase was ready for this to actually happen, the other ISPs stopped distributing their own special browsers entirely.
IE had won the browser wars so successfully that customers were using raw IE in spite of having the special ISP-specific version (of IE) available. So the dialup ISPs stopped doing that, and the broadband dealers had long-since known that people who go broadband have usually already gotten experience with the real browsers and avoided specialized software like the plague. This trend continued as the bugs and security holes of IE became known and the realization happened that one had to go use updated versions of the real IE to be *sure* you had a "fixed" version; with the ISP's version, you could never be sure of what was and wasn't fixed. The ISPs started running out of funds just trying to keep up with the security fixes Microsoft kept putting out all to support a dwindling userbase.
So in the end, why invest money maintaining a codebase of a hacked version of IE that's neither being used nor giving your customers any real value? And if not with IE, then there's nothing to be gained by doing it with Mozilla, even if it is "free" compared to licensing IE.
So the whole point of Mozilla as a means of developing capabilities for others to productize ended. nobody outside of Mozilla was really productizing it (the last straw really being when Apple went with the Konquerer baseline for Safari in spite of the speed improvements from 1.4 to 1.7), and Firefox has branded itself a hugely successful product in its own right.
thus, aside from maintaining a configuration UI that happens to work for a small subset of people, there's little to be gained from maintaining SeaMonkey as a released product. Gecko, Xul, and the other libraries will continue to improve to support Firefox and Thunderbird -- all that's missing is the use of a browser suite to show off their new features before going into Firefox.
and if its going into Firefox anyways, will anybody *really* miss it? The open-source philosophy will keep the Firefox people from writing generic features in such a way as to make it difficult to use them in other gecko-based products that are still out there or that will grow.
yeah, i saw it as an attempt to tax the internet without violating the federal ban on taxing the internet. of course, i didn't read the damned article, so i have no idea how scammers get into the act.
on that latter thing, its just a control factor, the illusion that "Everything will be better as long as *we* know who's doing what.". Total garbage, gross violation of the principles on which the nation was founded, but there you go.
as they are and what the future may hold are two different things. "quite a few and growing" is nice. but for the present, C#/.NET OSS is not nearly out there on the same scale as the other languages. all things (and all "absolute" statements) are relative.
of course, more OSS.NET would actually encourage more.NET development. its easier to get into a programming language/environment when you have (*a working*) something to use as either an example or as a baseline to fork from. text-book examples aren't nearly powerful enough to show the potential of a platform. (like how webservice examples suck because the main thing they demonstrate is a new way to get your stock-ticker).
i had reasonable success with Icecast on a slow linux box about 7 years ago. one thing about it was that at the time, the box would slow to a crawl if it had to re-encode the mp3 to a smaller value (ie., turning a 44/128 down into a 22/64). but keep in mind that was 7 years ago when boxes were 4 times slower (and far less memory) than today's hardware. shoutcast probably would have done the same, but I used icecast to stay on linux.
also, though broadcasting, i wasn't "advertising" per-se. i was using it mostly to share music with my then girlfriend (now wife) who was in jersey while i was here, and also to provide background music to a weekly chat i used to be part of before other commitments took out that weeknight.
but on the technical end with icecast and stuff, i'm sure things are better now than they were then. i just haven't had the need to try.
of course, the guy's only going to talk to Shareware developers. Shareware as a concept reflects Microsoft's #1 core value: people only write software to make money.
of course, the hundreds of opensource applications managed yearly simply don't exist to them because nobody make any money off of them.
"I've also tlkd to plenty of shareware developers and they certainly aren't using java. May use C/C++, Visual Basic, or Delphi." -- Fernandez.
*sigh*
of COURSE *shareware* developers are going to stick with windows. its the platform that created shareware, and has all the built-in tools (specifically, the registry) and the legacy of libraries out there to support the enforcement of shareware licenses by use-counts, disabled features, etc...
that and once you've had to pay for your development environment, of course you're going to want to get some money back for your products.
on the other hand, OpenSource client software exists in C, C++, Python, Java, Perl, and others, but until Mono gains a foothold (unlikely because skeptical developers fear when M$ pulls out the patent trump card they're holding), there will be very little C# open source out there. Free IDE support is out there because it, too, is open-source in many cases (eclipse being the major one).
professional developers use the right tool for the right job (theoretically). as such, we have pieces in vb (with vb.net add-ons leadng to an eventual full refactoring to vb.net) and java JFC, as necessary.
well, it used to be that setting up a track list and burning a cd was much harder than it became.
it takes first steps into something, actually having somebody do it at all, before the tools come out to make it easier to do.
already the tools are coming out to simplify the process by allowing you to edit mp3s (cuts, appending, etc) without having to turn everything into a wav (and go through a lossy re-encoding phase back to mp3). those tools weren't around 2 years ago.
in the end, it really is no more tedious to podcast than it is to make a mix tape or a mix cd.
collect the songs, record any voice-over segments you want, append them togther to a single mp3 stream, and stick it on your website, hoping there aren't so many downloads that it blows your monthly quota out.
what makes podcasting work is that you have to download it first. yes, it *seems* to those used to live radio that it sucks compared to streaming...but on the other hand, i can't stream 24/7, and you can't take a stream with you on the road.
i'd love to share my musical tastes by running my own online station...but at 34, with a day job and plenty of non-technical commitments, i can't. i'm not that kind of geek like i might have been in my college days. if i felt so inclined, casting like this would be a good alternative.
however, its also comparable to an audio-blog. you're mostly going to be interested in podcasts from people you already know.
and there are plenty of stations they'll air it. almost all of them are "talk" / "news" stations, or the regional traffic/weather stations above 210.
XM is trying for a different model -- the retail/restaurant PA system, to replace those stodgy (muzak) or over-the-air (with commercials) radio in the store. having commercial-free yet still "hip" is appealing to those places that want to keep the pop-current atmosphere and yet not want the commercials (or repetition) of commerical radio, and at a cheaper price than the old-style contracted music services that used dedicated phone lines or stuff like that.
it'll be a while before they get into the mall market that way, of course, but its a way to make more money (and advertising) without having to put commercials in the music stations directly.
annoyingly, the one major review of the myfi unit was from an mp3-o-phile who was so addicted to having total control that anything bad (like sound quality on some of the older rips) was a major bad.
as Fripp would say, the reviewer reviews himself far more than the target of his review.
the myfi isn't really my thing (the main time i listen with headphones is on the subway, where xm wouldn't reach. but one thing that will likely reduce my mp3-playing @ work would be listen.xmradio.com's online service.
i've not done the podcasting thing, though its really just a variation of making "mix tapes for friends" which became "making mixed cds for friends", which is now "making a mix as a giant mp3 stream for friends to download so i don't have to actually send them anything". its a kids thing, and if i was a kid now, i'd probably be in the heart of it.
at work, i tend to be in very tight control over what music is being played, both to control my mood (and ease frustrations when windows sucks) and to keep things inappropriate from being played when a strict-interpreter of the harrasment policy comes by (less they hear the wrong line of a Frank Zappa track).
but in my car things are mixed -- sometimes i don't want to "think" about what to put in (i've had times where i didn't play anything during the 30 minute drive home because i couldn't decide what to play...and it wasn't a quick decision: i was trying to decide what to play the entire drive home).
so there's a need for radio: when you don't have an mp3 collection in the car, it allows "randomness" to control what you hear. but as we all know, over the air radio sucks.
XM does have stations that cater to the kind of stuff I listen too. yes, they have a station that plays the "typical" classic rock, but hey, right now the 80s on 8 station is playing Live Aid complete (the full soundtrack to the 4-DVD set). They even "title" the segments so you know whether or not its coming from Wembley or Philly. What radio station has EVER played that since the day it aired back in 1985?
similarly, there are serveral "no rules" stations, like Fine Tuning, Deep Tracks, some of the alternative stations, an "unsigned" station, Cinemagic (which plays movie soundtracks intermixed with soundbytes), broadway, etc...
so unlike conventional radio where only one station in my area (DC) comes *close* to playing what I like, and that's the classic-rock staiton playing stuff so overplayed over the last 25 years that i'm sick of every single song they play, there are plenty of things to hear with XM.
The main place I have it on is in my kitchen while cooking dinner, 'cause its one of the places where i don't have time to "look" at anything, and i haven't wired it for playing mp3s from the main computer downstairs (and with XM, I probably won't).
I actually bought something because of hearing it on the radio for the first time in about 15 years -- the Lemony Snicket soundtrack, 'cause the closing-credits track is cool. Of course, that was XM radio.
agreed. this is particularly true at the local and state level where there are so many contradictions. some places let you parallel park "against traffic" (on the left side), others make that illegal. alcohol laws are rediculously variable (some places are dry, but you can bring your own booze into the restaurant and they'll provide glasses; yet other states bringing in booze into a place automatically gets you arrested on health-code violations and other laws). indecency and obscenity laws are utterly inconstant (as is the level at which the FCC does or doesn't enforce them). passing on the inside lane is illegal (yet never enforced unless its combined with a speeding incident to create an agressive driving case).
one form of common sense is that speeding is speeding and one should stay within the speed limit in spite of what the rest of traffic is doing around you.
yet another form of common sense says that driving below the general rate of travel is creating a road hazard, in spite of being at the legal limit.
different states may or may not give tickets to the slow driver in that instance. Britain, WILL ticket the slower driver. (Douglas Adams wrote that Graham Chapman used to keep both his British and his California license, and show each in the other country in order to plead ignorance for the cop and talk his way out of the ticket...that is, if he wasn't recognized as a celebraty, first).
when the law is concerned, there is little in common for common sense to have any meaning at all.
didn't see this post before i made my own, but i agree entirely. similarly, the credit card request is a specific form of credit check, a MUCH cheaper one than actually paying the credit agencies for the background check every time you want to do something like this (and if such a check happens to much, it can hurt your credit rating, too). its definitely better to let the credit card companies do that work and trust them to have done it right.
From the article: and arrived in rental cars that required a valid driver's license and one major credit card.
I'd argue that the rental car contract is not the same thing as flying without an ID.
A rental car company is libel when it gives a vehical away. It is required to be postively sure that the individual is fully capable and legally permitted to drive the car, or else it WILL be sued in the event of an accident, insurance be damned. Similarly, the credit card check is not to verify that the individual is what the ID says it is -- its to avoid having to do their own credit background check (minimum week or more delays and hefty increase in costs) in order to lend the car with the knowledge its going to someone likely to give it back.
they hold the credit card company responsible for dealing with that credit check and that cost, to save themselves the money and keep competitive.
both items are strictly business decisions that have no relation at all to the no flying without an ID law.
well, i don't doubt that if there WAS a microsoft player for linux, Xine and company would just recode their own tool to use the microsoft codec (and DRM code) for maximum compatibility over the reverse-engineered stuff. this is, in effect, what Xine did for Real, once the Real codec binary was released as part of the Helix code base.
still, in the end it all comes down to DRM, obnoxiously artificial limitations of the technology to keep the RIAA and MPAA happy (e.g., iTunes not being able to have more than one machine act as the "library" for an iPod). as long as linux has the "everything is free, even when it isn't" public perception, its an anethma to the entertainment industry's constant desire to increase control; as such, they'll never support a DRM scheme that includes linux support in it, so WMA and iTunes-protected AAC HAVE to stay proprietary and officially non-linux.
to support the stuff on linux is to imply that they're giving away the industry's "property" and to hollywood, they'd call that betrayal and your multi-million business model is over with in a heartbeat.
even though we know that just giving us the binaries doesn't mean we'll ALL use that to crack the codes and give it all away (like its not hard to make a "clean" digital master just by going down one analog generation or trapping the digital signal going to the sound card), hollywood won't see it that way. control is everything to them, in spite of the multi-billion dollar failures it breeds.
well, "linuxfund" would have to actually worry about specific usability issues. they would have to be willing to go to ALL of the distros and collect their various thoughts on what apps actually work (functionally) and what their customer complaints are about *how* they work (as far as layout and process to get something done goes).
this is precisely the kind of thing that happened with Firefox. they let mozilla-proper (seamonkey) deal with one aspect of the work (the security and standards-compliance issues of actually getting a browser to work at all, especially after the nightmare of the first Netscape source release...*ick*) and concentrated on what was mozilla's #1 flaw: an unintuitive UI oriented towards its target customer, being other programmers. making it look good and work well become the priority of Firefox, and in that, it became open-source's #4 success (behind BSD, apache, and linux, in that order).
OpenOffice already addressed such usability issues while it was still a proprietary product, competing (with microsoft) on usabiliy while playing catch-up on features. therefore, that doesn't count. OpenOffice's post-"free" work has mostly been in bug-fixes, format support, and portability. StarOffice *should* be looking at improving usability to surpass Office, but Sun doesn't have a clue about usability (if they did, Swing's "native" emulation would be a whole lot better than it is; Sun has always played catch-up in usability by trying to go their own way first (OpenLook vs. Motif, e.g.)).
So what else is there? GUI configuration tools for "LAMP" -- everybody's making them their own way, as part of their distro's "improvements over the core". with 60 reinventions of that wheel, they all suck. An independent company should work on that and all the linux distros contribute money and "use-cases" to make it good. Ditto with network configuration, printer configuration, etc. take the best one or two, combine them, and go from there.
*especially* stop making them dependent on loading a config file that was only configured with *their* tool. if the application can parse the file and make sense of it, the configuration tool should be able to do the same. if it can't deal 100% with every valid data entry in the config file, it sucks and get rid of it or fix it 'cause you'll always end up costing the customer time and money cleaning up the mess. (that's another clue -- treating your audience as clients and customers, even if they're not specifically paying you. they want value or they'll move on. and when you ARE being paid, don't consider the distro with the money to being the customer; you still code to your end user, not to the middleman with the checkbook, because that middleman wants paying end-users.)
Nautilus is far to heavy-weight to work with. a simple file manager should be among the lightest components in memory and cpu, not the heaviest (outside of firefox and any JVM, its the largest memory signature of any app on my box). Gnome needs to start over with that and trim it down...WAY down; externalize the lesser-used features into their own processes if necessary; that's what unix was designed for. Nautilus's usability still sucks, too.
the GTK and QT style guides should be more explicit, and programs that don't comply (*especially* with resizing windows...GOD i hate it when i make a window bigger and i get a ton of white-space) should be told they have a BUG and should fix it...and they should NOT be let into the distro or "desktop" until its fixed.
side rant follows:
trouble is, coding standards for gui apps never really got standardized reasonably. i didn't get into the open-source world in my motif days 'cause reading other people's motif (particularly those who knew less than i did, which was most) sucked. they hard-coded so much crap that it was impossible to modify reasonably. i didn't bother to get into GTK 'cause the problem was similar -- i was reading the same crappy kind of code, and didn't want to bother to lear
osx is not of a "dream" os, in that the apple gets in my way as much as windows does. i respect its achievements, but its not "done" either. (and i've 2 mac f[r]iends who've been trying to convert me for years. on the other hand, i don't see the point in paying apple's more-so-than-ever artificial price-point markups for their proprietary hardware and how, more so than windows, they're willing to say "screw backwards compatibility".
and quicktime and windows media for the apple suffer from the same usability problems as their windows counterparts (frequent upgrades of the whole interface just to stay codec compatible, and codecs changing for artifically-induced business "needs" rather than actual technological improvements), even if the video "looks better" on a mac (a subjective judgement at best).
yes, one is a "slave to the distro", but in the end, its either that or linux remain utterly the tool of the geek, not just those who are sick of microsofts vendor lockdowns. take your pick -- either linux usability improves to the point that issues like i've run into continually don't happen, or linux remains the tool of "intellectual elites" using it as a way of saying "i'm better than you 'cause i know this".
one or the other, but linux can't serve as both, and its use at the latter isn't really solving peoples real computing problems.
again, start looking at life from a usability standpoint and not a "well it can be done if you know how, therefore its fine" viewpoint, and things REALLY change. not just computers, but all of life itself. Go read "The Design of Ordinary Things" (likely in the science and technology center of borders / bn).
we did things one way out of need. that need is gone, so we shouldn't have to do things that way anymore.
open source is still critical, especially for products in their early stages and for those that need to manage optimizations and bug fixes without turnaround times (i.e., distro makers and ISPs and the like). but the effects of opensource (rapid bug-fixes in immature products, high security and robustness levels) should be felt by all without the inherent responsibility of actually having to compile source code, once products reach a reasonable level of robustness.
linux media players have had 10 years since awt-tv to get that way. they still haven't. every time a product started to get somewhere, the developers abandoned it for the next big thing, still in source-code/alpha stages. OR they abandoned it for a new pet project or actually doing their day job. that pattern repeated itself for years. its time to stop, pick an app, make it usable (by the standards i've defined in the previous post) and be done with it, rather than move on to a new framework 'cause you got bored with the old one. (that in itself IS a valid open-source negative: products survive only as long as someone feels like maintaining them).
gtk alone has had at least 6 media players since the library's 1.0, plus music and movie players that used gtk for menus while having their own proprietary look&feel (xmms, e.g.). some players claim to be browser plug-ins while in reality force the movie into a separate app (bad usability juju). and *still* don't work right in that form.
there are better ways to do it, but i don't have the time or resources to write (or manage) an opensource project to do it the right way.
i think you don't know who/what i am and where i'm coming from.
look at my slashdot ID number (hint: it has 2 fewer *digits* than yours). i was running open-source software on Suns, decstations on ultrix, aix boxen, data general, minix, and (god what a pain in the arse) variations of system five including SCO and Esix for 386es, plus VMS and pre-windows DOS, all more than a decade before "Open Source" became a term and GNU was only known for its compiler, emacs (which was a usability nightmare before we even knew what THAT term meant), and a pipe-dream of an OS that was known to be unachievable until some upstart Finnish dude actually achieved it before they did.
for all of my college existence and more than half my professional life, "compile your own damn version" was all we had. i've lived that life, and i don't see the point in living it anymore.
now, i'm the usability specialist at my company (the only one) because i'm willing to flat out say "we don't need to do shit like this anymore." and just as i'm willing to say it for my company's own products, i'm willing to say it about certain ways of working with linux as well as windows.
"just compile from the source and all will be ok" is no longer the way to do things. its the right way for distro managers to manage distros, but not the way to distribute products for an end user anymore.
A video player is an end-user product. it should work "out of the box", not be dependent on obscure libraries (or always run latest-and-greatest versions of said libraries), be near-totally transpearant in looking up codecs its missing (no, windows products don't do that right either), and not upgrade the core system except for significant usability fixes (meaning eventually, stop upgrading 'cause it should just look good and work right...again, windows products suck at this regard just as much as linux, especially when they're used to "preview" the next OS look&feel, which both quicktime and windows player are infamous for).
as i said, invariably upgrading the player usually has involved upgrading the version of gtk or some other dependent library, which involves upgrading another, and/or pre-built binaries aren't compatible with my version of libc or libc++ simply because i'm a year out of date (outside of security fixes) and i have too many other things to do that need a stable, unchanging system to worry about upgrading my box every 6 months to keep up with a video player i barely use. i don't WANT to have to upgrade my distro, 'cause for my programming purposes (100% java), its FINE the way it is and i don't want to muck with that. latest and greatest distro likely will include software that will put a load on my (old) cpu and relatively low memory (firefox alone is heavier than anything thrown on that box before it besides running 2 jvms).
i also am sick and tired of the "just recompile it" crap. we should be past that now. i lived that life in the 90s when it was the only way to do anything at all, and now, i'm sick of it. i'm not that kind of geek anymore; i've got better things to do with my time. this is what happens to geeks when they cross age 30, get married, and find things outside of computers to actually be involved in.
its like my being "out of the loop" with regards to video cards and crap like that -- i don't play games so i don't need the "latest and greatest" hardware that the "latest and greatest" games require.
i'll live without upgrading my box for mplayer or whatever. its as simple as that. i don't need it that badly (well, at all, really). i have an XP box i use specifically for multi-media and little else.
yes, windows doesn't work "the best", but it works enough, "out of the box" for what i need. it sucks as a development platform for java, but its what i'm stuck with at work (though i have a linux box at work too, specifically for web development on a LAMP system 'cause that's one place linux WILL always be better than windows at). yes, i tire of having to upgrade windows media or quicktime 'cause they like changing their codecs every 6 months when crackers reverse-engineer them, but at least in windows its "one-click" (+ agreeing to sell your firstborn to Microsoft in the EULA).
its not "upgrade the player by searching for the damned rpm, finding out i need 3 other rpms to get it to work, one of which doesn't exist for my particular dist because of an incompatible library rpm that was upgraded only in the next full version of the distro entirely".
its not FUD -- it is my specific experiences with every video player i've ever installed from redhat 6 to fedora core 2, and i'm not in the mood to "keep upgrading and all will be better" any more. i don't have time to blow an afternoon upgrading my box, finding out i need more memory, or what the hell ever. i'm tired of that lifestyle. i know that there are better distros for desktop/media work, but they get in my way as development platforms.
and it still doesn't make my statement about the legality of WMA on linux FUD at all. codecs to view protected AAC or WMA on unlicensed software are technically illegal by virtue of the DMCA. its bullshit, and i'm the first to support those willing to be consciencious objectors engaging in civil disobedience.
but for myself, i have more important and personal battles to fight.
Slashdot Mirror
Don't assume that I was arguing in favor of things being different than they are as such. I am not making decisions, merely presenting opinions and observations (that are more valid than you imply here).
I never said NASA was disfunctional in any way, though being a bureacracy (strike one) made up of humans (strike two) doesn't exactly inspire an high degree of confidence. But I feel that way with most human bureaucracies, so I learn to not worry about it. Certainly the (almost) perfect Cassini-Huygens work so far has made up greatly for many of the human errors of the Hubble first-try and Mars Probes, and the great unknown error in the Galileo probe's antenna.
I was not recommending that NASA become privatized, only that they private companies think ahead a little better and PLAN so that when things have to be replaced, the replacements are ready. I was not specifically saying that Hubble needs to be replaced ASAP (though I certainly would like it to, and I'm *hardly* alone on that, given what I've gleaned from friends who work in NASA). I'm just saying that they had nothing on the table for that eventuality, just as they currently have little on the table for the eventuality of eliminating the shuttle fleet.
That was all my comparison was meant to make, and I said nothing as such about desiring to privatize NASA and my follow-up post did mention the "profit" nature and how long-term research doesn't lead to profit in a recognizable sense.
Most companies, when they see a success, *immediately* begin plans for the next version. Software, entertainment, "inventions". They get a tangible benefit from something, see places where it might be better, and get started on building the next while another group continues to get what they can supporting the old.
NASA has done this before, but there are times when they act like they've never heard of this concept, and Hubble's fate to me is one of those times. Hubble dies in 2008, and the best chance we've been told for a replacement is 2020 (last I heard). THAT is lousy planning for a program most researchers consider a great success (once they got past the initial technical goof).
Granted, the crappy budget the Feds give them doesn't exactly help, being the result of the fact that Washington is just as short-sighted as Wall Street is, as demonstrated by the DARPA decision posted elsewhere on this site.
Planned obsolescense is an entirely different thing. It's not about making a better product, it's about making crappy products that break so they can sell you a replacement that doesn't do anything better than the old one.
yes, but it came about under the realization that even building for "the long term" only puts you out of business (you run out of customers as everybody who wants one has one), or people buy a new one anyways while their old one still works, just for aesthetic reasons.
No matter how it is, the fact that p.o. is practiced is a result of companies realizing, regardless of particular profit motives, that you have to have a new model ready before the old one is really obsolete, to keep the cash flow coming.
NASA should look at their stuff the same way -- only instead of cash, think "information". Hubble is going to stop sending data, and there's nothing going up there to send the same type of data back. The shuttles are going to be decomissioned, and there is nothing in the production stages that will be ready to take their place when it happens. The information stops, because they didn't plan enough on what to do when the suppliers of that information that seemed briefly so unlimited stopped functioning permanently.
I agree that *nobody* treats long-term research with any respect anymore, thanks to Wall Street's permanent emphasis on "quarterly results" at the expense of a company's real future. The final result of all this seems to be a world created *entirely* by amateurs.
acknowledging, of course, that as a commercial enterprise, NASA would have to find some way (besides selling satellite launch services) to actually turn a profit...and the curse is that long-term research isn't *meant* to turn a profit "now".
yeah. the *real* problem with NASA is they see these end-of-usefulness deadlines as being N years off, and forget that any replacement technology needs N+M years to develop, but because N years seems so far away, they don't start on the replacement...
so here we are with a shuttle fleet on the virge of permanent decommision (and 2 lost already) and a hubble satellite, and no plans for an actual replacement because N years hit NASA a LOT sooner than they realized...
Its the problem of NASA not being a commercial enterprise. The IT world, the auto world, the airplane world, the appliance world, the electronics world have all lived under the spectre of Plan Obsolescense for decades. We *know* that we have to have the replacement for something read *before* the current model goes out of usefulness.
NASA has never lived under that problem before, so as an institution, it simply didn't know how to react.
yeah, it really comes down to question of "what would you do with it". its all good and well that everybody likes feature "X" that comes from the integrated suite that currently the separated apps can't give (doesn't mean they won't get some form of triggering the other apps as a result of this announcement), but what more do you want besides a browser that stays the way you already know?
Mozilla SeaMonkey drove producing a set of capabilities cloning the original Netscape base it derived from (in a very detached way, of course), with the idea that Netscape/AOL (and others) could take the baseline and produce *standards-compliant* browsers on top of those capabilities.
3 things happened, and one thing *didn't* happen.
1) Netscape 6.x and 7.x were successfully rolled out based on the Mozilla baseline (only now 8.x has already started coming from the Firefox base, and the "Communicator" concept is gone)
2) AOL decided, in spite of their investment, to give up the idea of actually doing an AOL browser based on Mozilla in favor of playing marketting games with Microsoft by supporting IE instead in exchange for not being blacklisted off of Microsoft's illegal (but still practiced) OEM deals.
3) Firefox came out and had a marketting push unlike any other open source project around, including Linux distros themselves. And to top it off, the damn thing actually works (those few slashdotter complaints in this thread notwithstanding).
the thing that didn't happen: other ISPs didn't build their own browsers on Mozilla tech. It used to be in the 90s that all the ISPs followed AOL and Compuserve's lead in taking an existing browser (usually IE) and specializing it to become part of their service. Mozilla was setting up its code base specifically for that purpose -- we provide standards-compliance and ease-of-skinning; you skin it to meet your customer's needs. Only by the time Mozilla's codebase was ready for this to actually happen, the other ISPs stopped distributing their own special browsers entirely.
IE had won the browser wars so successfully that customers were using raw IE in spite of having the special ISP-specific version (of IE) available. So the dialup ISPs stopped doing that, and the broadband dealers had long-since known that people who go broadband have usually already gotten experience with the real browsers and avoided specialized software like the plague. This trend continued as the bugs and security holes of IE became known and the realization happened that one had to go use updated versions of the real IE to be *sure* you had a "fixed" version; with the ISP's version, you could never be sure of what was and wasn't fixed. The ISPs started running out of funds just trying to keep up with the security fixes Microsoft kept putting out all to support a dwindling userbase.
So in the end, why invest money maintaining a codebase of a hacked version of IE that's neither being used nor giving your customers any real value? And if not with IE, then there's nothing to be gained by doing it with Mozilla, even if it is "free" compared to licensing IE.
So the whole point of Mozilla as a means of developing capabilities for others to productize ended. nobody outside of Mozilla was really productizing it (the last straw really being when Apple went with the Konquerer baseline for Safari in spite of the speed improvements from 1.4 to 1.7), and Firefox has branded itself a hugely successful product in its own right.
thus, aside from maintaining a configuration UI that happens to work for a small subset of people, there's little to be gained from maintaining SeaMonkey as a released product. Gecko, Xul, and the other libraries will continue to improve to support Firefox and Thunderbird -- all that's missing is the use of a browser suite to show off their new features before going into Firefox.
and if its going into Firefox anyways, will anybody *really* miss it? The open-source philosophy will keep the Firefox people from writing generic features in such a way as to make it difficult to use them in other gecko-based products that are still out there or that will grow.
yeah, i saw it as an attempt to tax the internet without violating the federal ban on taxing the internet. of course, i didn't read the damned article, so i have no idea how scammers get into the act.
on that latter thing, its just a control factor, the illusion that "Everything will be better as long as *we* know who's doing what.". Total garbage, gross violation of the principles on which the nation was founded, but there you go.
as they are and what the future may hold are two different things. "quite a few and growing" is nice. but for the present, C#/.NET OSS is not nearly out there on the same scale as the other languages. all things (and all "absolute" statements) are relative.
.NET would actually encourage more .NET development. its easier to get into a programming language/environment when you have (*a working*) something to use as either an example or as a baseline to fork from. text-book examples aren't nearly powerful enough to show the potential of a platform. (like how webservice examples suck because the main thing they demonstrate is a new way to get your stock-ticker).
of course, more OSS
i had reasonable success with Icecast on a slow linux box about 7 years ago. one thing about it was that at the time, the box would slow to a crawl if it had to re-encode the mp3 to a smaller value (ie., turning a 44/128 down into a 22/64). but keep in mind that was 7 years ago when boxes were 4 times slower (and far less memory) than today's hardware. shoutcast probably would have done the same, but I used icecast to stay on linux.
also, though broadcasting, i wasn't "advertising" per-se. i was using it mostly to share music with my then girlfriend (now wife) who was in jersey while i was here, and also to provide background music to a weekly chat i used to be part of before other commitments took out that weeknight.
but on the technical end with icecast and stuff, i'm sure things are better now than they were then. i just haven't had the need to try.
and pardon me while i *duh* myself.
of course, the guy's only going to talk to Shareware developers. Shareware as a concept reflects Microsoft's #1 core value: people only write software to make money.
of course, the hundreds of opensource applications managed yearly simply don't exist to them because nobody make any money off of them.
"I've also tlkd to plenty of shareware developers and they certainly aren't using java. May use C/C++, Visual Basic, or Delphi." -- Fernandez.
*sigh*
of COURSE *shareware* developers are going to stick with windows. its the platform that created shareware, and has all the built-in tools (specifically, the registry) and the legacy of libraries out there to support the enforcement of shareware licenses by use-counts, disabled features, etc...
that and once you've had to pay for your development environment, of course you're going to want to get some money back for your products.
on the other hand, OpenSource client software exists in C, C++, Python, Java, Perl, and others, but until Mono gains a foothold (unlikely because skeptical developers fear when M$ pulls out the patent trump card they're holding), there will be very little C# open source out there. Free IDE support is out there because it, too, is open-source in many cases (eclipse being the major one).
professional developers use the right tool for the right job (theoretically). as such, we have pieces in vb (with vb.net add-ons leadng to an eventual full refactoring to vb.net) and java JFC, as necessary.
well, it used to be that setting up a track list and burning a cd was much harder than it became.
it takes first steps into something, actually having somebody do it at all, before the tools come out to make it easier to do.
already the tools are coming out to simplify the process by allowing you to edit mp3s (cuts, appending, etc) without having to turn everything into a wav (and go through a lossy re-encoding phase back to mp3). those tools weren't around 2 years ago.
in the end, it really is no more tedious to podcast than it is to make a mix tape or a mix cd.
collect the songs, record any voice-over segments you want, append them togther to a single mp3 stream, and stick it on your website, hoping there aren't so many downloads that it blows your monthly quota out.
what makes podcasting work is that you have to download it first. yes, it *seems* to those used to live radio that it sucks compared to streaming...but on the other hand, i can't stream 24/7, and you can't take a stream with you on the road.
i'd love to share my musical tastes by running my own online station...but at 34, with a day job and plenty of non-technical commitments, i can't. i'm not that kind of geek like i might have been in my college days. if i felt so inclined, casting like this would be a good alternative.
however, its also comparable to an audio-blog. you're mostly going to be interested in podcasts from people you already know.
and there are plenty of stations they'll air it. almost all of them are "talk" / "news" stations, or the regional traffic/weather stations above 210.
XM is trying for a different model -- the retail/restaurant PA system, to replace those stodgy (muzak) or over-the-air (with commercials) radio in the store. having commercial-free yet still "hip" is appealing to those places that want to keep the pop-current atmosphere and yet not want the commercials (or repetition) of commerical radio, and at a cheaper price than the old-style contracted music services that used dedicated phone lines or stuff like that.
it'll be a while before they get into the mall market that way, of course, but its a way to make more money (and advertising) without having to put commercials in the music stations directly.
annoyingly, the one major review of the myfi unit was from an mp3-o-phile who was so addicted to having total control that anything bad (like sound quality on some of the older rips) was a major bad.
as Fripp would say, the reviewer reviews himself far more than the target of his review.
the myfi isn't really my thing (the main time i listen with headphones is on the subway, where xm wouldn't reach. but one thing that will likely reduce my mp3-playing @ work would be listen.xmradio.com's online service.
i've not done the podcasting thing, though its really just a variation of making "mix tapes for friends" which became "making mixed cds for friends", which is now "making a mix as a giant mp3 stream for friends to download so i don't have to actually send them anything". its a kids thing, and if i was a kid now, i'd probably be in the heart of it.
at work, i tend to be in very tight control over what music is being played, both to control my mood (and ease frustrations when windows sucks) and to keep things inappropriate from being played when a strict-interpreter of the harrasment policy comes by (less they hear the wrong line of a Frank Zappa track).
but in my car things are mixed -- sometimes i don't want to "think" about what to put in (i've had times where i didn't play anything during the 30 minute drive home because i couldn't decide what to play...and it wasn't a quick decision: i was trying to decide what to play the entire drive home).
so there's a need for radio: when you don't have an mp3 collection in the car, it allows "randomness" to control what you hear. but as we all know, over the air radio sucks.
XM does have stations that cater to the kind of stuff I listen too. yes, they have a station that plays the "typical" classic rock, but hey, right now the 80s on 8 station is playing Live Aid complete (the full soundtrack to the 4-DVD set). They even "title" the segments so you know whether or not its coming from Wembley or Philly. What radio station has EVER played that since the day it aired back in 1985?
similarly, there are serveral "no rules" stations, like Fine Tuning, Deep Tracks, some of the alternative stations, an "unsigned" station, Cinemagic (which plays movie soundtracks intermixed with soundbytes), broadway, etc...
so unlike conventional radio where only one station in my area (DC) comes *close* to playing what I like, and that's the classic-rock staiton playing stuff so overplayed over the last 25 years that i'm sick of every single song they play, there are plenty of things to hear with XM.
The main place I have it on is in my kitchen while cooking dinner, 'cause its one of the places where i don't have time to "look" at anything, and i haven't wired it for playing mp3s from the main computer downstairs (and with XM, I probably won't).
I actually bought something because of hearing it on the radio for the first time in about 15 years -- the Lemony Snicket soundtrack, 'cause the closing-credits track is cool. Of course, that was XM radio.
already pointed out (and i do that mistake constantly), but thanks :)
i make that mistake constantly, and this got pointed out in my blog entry on the subject so i was kinda expecting it to hit here too...
*sigh*
agreed. this is particularly true at the local and state level where there are so many contradictions. some places let you parallel park "against traffic" (on the left side), others make that illegal. alcohol laws are rediculously variable (some places are dry, but you can bring your own booze into the restaurant and they'll provide glasses; yet other states bringing in booze into a place automatically gets you arrested on health-code violations and other laws). indecency and obscenity laws are utterly inconstant (as is the level at which the FCC does or doesn't enforce them). passing on the inside lane is illegal (yet never enforced unless its combined with a speeding incident to create an agressive driving case).
one form of common sense is that speeding is speeding and one should stay within the speed limit in spite of what the rest of traffic is doing around you.
yet another form of common sense says that driving below the general rate of travel is creating a road hazard, in spite of being at the legal limit.
different states may or may not give tickets to the slow driver in that instance. Britain, WILL ticket the slower driver. (Douglas Adams wrote that Graham Chapman used to keep both his British and his California license, and show each in the other country in order to plead ignorance for the cop and talk his way out of the ticket...that is, if he wasn't recognized as a celebraty, first).
when the law is concerned, there is little in common for common sense to have any meaning at all.
didn't see this post before i made my own, but i agree entirely. similarly, the credit card request is a specific form of credit check, a MUCH cheaper one than actually paying the credit agencies for the background check every time you want to do something like this (and if such a check happens to much, it can hurt your credit rating, too). its definitely better to let the credit card companies do that work and trust them to have done it right.
From the article: and arrived in rental cars that required a valid driver's license and one major credit card.
I'd argue that the rental car contract is not the same thing as flying without an ID.
A rental car company is libel when it gives a vehical away. It is required to be postively sure that the individual is fully capable and legally permitted to drive the car, or else it WILL be sued in the event of an accident, insurance be damned. Similarly, the credit card check is not to verify that the individual is what the ID says it is -- its to avoid having to do their own credit background check (minimum week or more delays and hefty increase in costs) in order to lend the car with the knowledge its going to someone likely to give it back.
they hold the credit card company responsible for dealing with that credit check and that cost, to save themselves the money and keep competitive.
both items are strictly business decisions that have no relation at all to the no flying without an ID law.
well, i don't doubt that if there WAS a microsoft player for linux, Xine and company would just recode their own tool to use the microsoft codec (and DRM code) for maximum compatibility over the reverse-engineered stuff. this is, in effect, what Xine did for Real, once the Real codec binary was released as part of the Helix code base.
still, in the end it all comes down to DRM, obnoxiously artificial limitations of the technology to keep the RIAA and MPAA happy (e.g., iTunes not being able to have more than one machine act as the "library" for an iPod). as long as linux has the "everything is free, even when it isn't" public perception, its an anethma to the entertainment industry's constant desire to increase control; as such, they'll never support a DRM scheme that includes linux support in it, so WMA and iTunes-protected AAC HAVE to stay proprietary and officially non-linux.
to support the stuff on linux is to imply that they're giving away the industry's "property" and to hollywood, they'd call that betrayal and your multi-million business model is over with in a heartbeat.
even though we know that just giving us the binaries doesn't mean we'll ALL use that to crack the codes and give it all away (like its not hard to make a "clean" digital master just by going down one analog generation or trapping the digital signal going to the sound card), hollywood won't see it that way. control is everything to them, in spite of the multi-billion dollar failures it breeds.
looking back, i think the fact that apple didn't call their bluff is the main reason M$ didn't bother to actually go through with it.
well, "linuxfund" would have to actually worry about specific usability issues. they would have to be willing to go to ALL of the distros and collect their various thoughts on what apps actually work (functionally) and what their customer complaints are about *how* they work (as far as layout and process to get something done goes).
this is precisely the kind of thing that happened with Firefox. they let mozilla-proper (seamonkey) deal with one aspect of the work (the security and standards-compliance issues of actually getting a browser to work at all, especially after the nightmare of the first Netscape source release...*ick*) and concentrated on what was mozilla's #1 flaw: an unintuitive UI oriented towards its target customer, being other programmers. making it look good and work well become the priority of Firefox, and in that, it became open-source's #4 success (behind BSD, apache, and linux, in that order).
OpenOffice already addressed such usability issues while it was still a proprietary product, competing (with microsoft) on usabiliy while playing catch-up on features. therefore, that doesn't count. OpenOffice's post-"free" work has mostly been in bug-fixes, format support, and portability. StarOffice *should* be looking at improving usability to surpass Office, but Sun doesn't have a clue about usability (if they did, Swing's "native" emulation would be a whole lot better than it is; Sun has always played catch-up in usability by trying to go their own way first (OpenLook vs. Motif, e.g.)).
So what else is there? GUI configuration tools for "LAMP" -- everybody's making them their own way, as part of their distro's "improvements over the core". with 60 reinventions of that wheel, they all suck. An independent company should work on that and all the linux distros contribute money and "use-cases" to make it good. Ditto with network configuration, printer configuration, etc. take the best one or two, combine them, and go from there.
*especially* stop making them dependent on loading a config file that was only configured with *their* tool. if the application can parse the file and make sense of it, the configuration tool should be able to do the same. if it can't deal 100% with every valid data entry in the config file, it sucks and get rid of it or fix it 'cause you'll always end up costing the customer time and money cleaning up the mess. (that's another clue -- treating your audience as clients and customers, even if they're not specifically paying you. they want value or they'll move on. and when you ARE being paid, don't consider the distro with the money to being the customer; you still code to your end user, not to the middleman with the checkbook, because that middleman wants paying end-users.)
Nautilus is far to heavy-weight to work with. a simple file manager should be among the lightest components in memory and cpu, not the heaviest (outside of firefox and any JVM, its the largest memory signature of any app on my box). Gnome needs to start over with that and trim it down...WAY down; externalize the lesser-used features into their own processes if necessary; that's what unix was designed for. Nautilus's usability still sucks, too.
the GTK and QT style guides should be more explicit, and programs that don't comply (*especially* with resizing windows...GOD i hate it when i make a window bigger and i get a ton of white-space) should be told they have a BUG and should fix it...and they should NOT be let into the distro or "desktop" until its fixed.
side rant follows:
trouble is, coding standards for gui apps never really got standardized reasonably. i didn't get into the open-source world in my motif days 'cause reading other people's motif (particularly those who knew less than i did, which was most) sucked. they hard-coded so much crap that it was impossible to modify reasonably. i didn't bother to get into GTK 'cause the problem was similar -- i was reading the same crappy kind of code, and didn't want to bother to lear
osx is not of a "dream" os, in that the apple gets in my way as much as windows does. i respect its achievements, but its not "done" either. (and i've 2 mac f[r]iends who've been trying to convert me for years. on the other hand, i don't see the point in paying apple's more-so-than-ever artificial price-point markups for their proprietary hardware and how, more so than windows, they're willing to say "screw backwards compatibility".
and quicktime and windows media for the apple suffer from the same usability problems as their windows counterparts (frequent upgrades of the whole interface just to stay codec compatible, and codecs changing for artifically-induced business "needs" rather than actual technological improvements), even if the video "looks better" on a mac (a subjective judgement at best).
yes, one is a "slave to the distro", but in the end, its either that or linux remain utterly the tool of the geek, not just those who are sick of microsofts vendor lockdowns. take your pick -- either linux usability improves to the point that issues like i've run into continually don't happen, or linux remains the tool of "intellectual elites" using it as a way of saying "i'm better than you 'cause i know this".
one or the other, but linux can't serve as both, and its use at the latter isn't really solving peoples real computing problems.
again, start looking at life from a usability standpoint and not a "well it can be done if you know how, therefore its fine" viewpoint, and things REALLY change. not just computers, but all of life itself. Go read "The Design of Ordinary Things" (likely in the science and technology center of borders / bn).
we did things one way out of need. that need is gone, so we shouldn't have to do things that way anymore.
open source is still critical, especially for products in their early stages and for those that need to manage optimizations and bug fixes without turnaround times (i.e., distro makers and ISPs and the like). but the effects of opensource (rapid bug-fixes in immature products, high security and robustness levels) should be felt by all without the inherent responsibility of actually having to compile source code, once products reach a reasonable level of robustness.
linux media players have had 10 years since awt-tv to get that way. they still haven't. every time a product started to get somewhere, the developers abandoned it for the next big thing, still in source-code/alpha stages. OR they abandoned it for a new pet project or actually doing their day job. that pattern repeated itself for years. its time to stop, pick an app, make it usable (by the standards i've defined in the previous post) and be done with it, rather than move on to a new framework 'cause you got bored with the old one. (that in itself IS a valid open-source negative: products survive only as long as someone feels like maintaining them).
gtk alone has had at least 6 media players since the library's 1.0, plus music and movie players that used gtk for menus while having their own proprietary look&feel (xmms, e.g.). some players claim to be browser plug-ins while in reality force the movie into a separate app (bad usability juju). and *still* don't work right in that form.
there are better ways to do it, but i don't have the time or resources to write (or manage) an opensource project to do it the right way.
i think you don't know who/what i am and where i'm coming from.
look at my slashdot ID number (hint: it has 2 fewer *digits* than yours). i was running open-source software on Suns, decstations on ultrix, aix boxen, data general, minix, and (god what a pain in the arse) variations of system five including SCO and Esix for 386es, plus VMS and pre-windows DOS, all more than a decade before "Open Source" became a term and GNU was only known for its compiler, emacs (which was a usability nightmare before we even knew what THAT term meant), and a pipe-dream of an OS that was known to be unachievable until some upstart Finnish dude actually achieved it before they did.
for all of my college existence and more than half my professional life, "compile your own damn version" was all we had. i've lived that life, and i don't see the point in living it anymore.
now, i'm the usability specialist at my company (the only one) because i'm willing to flat out say "we don't need to do shit like this anymore." and just as i'm willing to say it for my company's own products, i'm willing to say it about certain ways of working with linux as well as windows.
"just compile from the source and all will be ok" is no longer the way to do things. its the right way for distro managers to manage distros, but not the way to distribute products for an end user anymore.
A video player is an end-user product. it should work "out of the box", not be dependent on obscure libraries (or always run latest-and-greatest versions of said libraries), be near-totally transpearant in looking up codecs its missing (no, windows products don't do that right either), and not upgrade the core system except for significant usability fixes (meaning eventually, stop upgrading 'cause it should just look good and work right...again, windows products suck at this regard just as much as linux, especially when they're used to "preview" the next OS look&feel, which both quicktime and windows player are infamous for).
as i said, invariably upgrading the player usually has involved upgrading the version of gtk or some other dependent library, which involves upgrading another, and/or pre-built binaries aren't compatible with my version of libc or libc++ simply because i'm a year out of date (outside of security fixes) and i have too many other things to do that need a stable, unchanging system to worry about upgrading my box every 6 months to keep up with a video player i barely use. i don't WANT to have to upgrade my distro, 'cause for my programming purposes (100% java), its FINE the way it is and i don't want to muck with that. latest and greatest distro likely will include software that will put a load on my (old) cpu and relatively low memory (firefox alone is heavier than anything thrown on that box before it besides running 2 jvms).
i also am sick and tired of the "just recompile it" crap. we should be past that now. i lived that life in the 90s when it was the only way to do anything at all, and now, i'm sick of it. i'm not that kind of geek anymore; i've got better things to do with my time. this is what happens to geeks when they cross age 30, get married, and find things outside of computers to actually be involved in.
its like my being "out of the loop" with regards to video cards and crap like that -- i don't play games so i don't need the "latest and greatest" hardware that the "latest and greatest" games require.
i'll live without upgrading my box for mplayer or whatever. its as simple as that. i don't need it that badly (well, at all, really). i have an XP box i use specifically for multi-media and little else.
yes, windows doesn't work "the best", but it works enough, "out of the box" for what i need. it sucks as a development platform for java, but its what i'm stuck with at work (though i have a linux box at work too, specifically for web development on a LAMP system 'cause that's one place linux WILL always be better than windows at). yes, i tire of having to upgrade windows media or quicktime 'cause they like changing their codecs every 6 months when crackers reverse-engineer them, but at least in windows its "one-click" (+ agreeing to sell your firstborn to Microsoft in the EULA).
its not "upgrade the player by searching for the damned rpm, finding out i need 3 other rpms to get it to work, one of which doesn't exist for my particular dist because of an incompatible library rpm that was upgraded only in the next full version of the distro entirely".
its not FUD -- it is my specific experiences with every video player i've ever installed from redhat 6 to fedora core 2, and i'm not in the mood to "keep upgrading and all will be better" any more. i don't have time to blow an afternoon upgrading my box, finding out i need more memory, or what the hell ever. i'm tired of that lifestyle. i know that there are better distros for desktop/media work, but they get in my way as development platforms.
and it still doesn't make my statement about the legality of WMA on linux FUD at all. codecs to view protected AAC or WMA on unlicensed software are technically illegal by virtue of the DMCA. its bullshit, and i'm the first to support those willing to be consciencious objectors engaging in civil disobedience.
but for myself, i have more important and personal battles to fight.