in which developers create free intellectual property only to have others scoop it up and generate huge amounts of revenue
The only way* for a company to make "huge amounts of revenue" from Open Source software is to add value so that people are prepared to pay you money for something that they could get elsewhere for free. That "value" might be providing top quality support, or it might be investigating in marketing or just having a number of employees who wear suits and use words like "leverage" that give corporate clients a warm fuzzy feeling. Either way, does anybody really have a problem with that?
Any company director who looses sleep about getting all this "money for nothing" simply needs to let their employees use some of their paid time to contribute to writing OSS code or coordinating OSS development.
*(excluding the "extort protection money on the back of questionable IP violation claims" method, of course).
Not only is it an urban legend. Fact of the matter is it is more wise to use pressurized pens in space. That's because pencils contain 'graphite' with is conductive. If graphite dust infiltrated sensitive electronics, it could be disastrous.
But inside every pencil is an inanimate carbon rod and I'm sure we all know what a vital resource that can be in space travel.
PS: I am fully cogniscent that accounts of NASA's profligate spending on microgravity-enabled ballpoint pens are widely disputed and most probably false - however, the widespread recognition of the abovementioned meme facilitated the employment of a comedic device known as a "double entendre" in which apparently innocent phrases quoted in a suitable context become amenable to an alternative interpretation of a sexual, or otherwise taboo, nature, to humourous effect. Also, for the further avoidance of doubt, I do not have reliable evidence for the insinuation that Russian cosmonauts make improper use of writing implements or, indeed, that said implements can be satisfactorally used for auto-erotic purposes. Finally, I believe that the assertion that all Slashdot readers are celebate, if not virgin, is unjustified and that many have, in fact, experienced sexual relations - sometimes even with other people.
I apologise profusely for neglecting to negate whatever modest humour my original posting may have possessed by introducing the above caveats at an earlier stage in proceedings.
"One topic that is evidently too hot to handle: How do you cope with sexual desire among healthy young men and women during a mission years long?"
A: Spend $100,000,000 developing high tech, er, appliances that work in zero gravity, then brace for the ensuing scandal when it emerges that the Russians just used pencils...
Alternatively, recruit more nerds and less jocks. Why not advertise on Slashdot?
But if you actually take a look at the 1.3 billion phones that get sold, I'd prefer to have our software in 60% or 70% or 80% of them, than I would to have 2% or 3%, which is what Apple might get.
...but iPhone and Windows Mobile aren't competing with the 1.3 billion phones sold. They are competing for the same, relatively small subset of the market representing high-end smartphones with EMAIL and serious web-browsing facilities.
Now, in the MS corner, we have Windows Mobile. I have a fairly high-end Windows smartphone, and while I like the pwer of the thing, using it can be like kicking a dead whale along a beach. Using it as a music player is particularly excruciating. I would be very reluctant to recommend it to a non techie. Mind you, some basic phones are pretty nasty to use, as well.
Although nobody has seen an iPhone properly, Apple have a track record and would need to be having a very bad day not to produce something vastly more usable. Its not like XP vs. OSX: personally, ignoring security and reliability, I don't think that there is any clear blue water between OSX and Win2K/XP on the usability front - I actually prefer the XP GUI in some ways - but Windows Mobile feels like a throwback to Windows 2.
Its not that Apple stuff is perfect - just that it usually gives the impression that it was actually designed by people who gave a damn about the product in a world where many - if not most - other computer and home electronics products seem to have congealed out of a committee process.
Also - Apple seem to be capable of making "less is more" decisions. Notice that hardly anybody has matched Apple's minimalist design style? Other manufacturers have produced designer-y ranges but the extra buttons, chrome grills, go-faster stripes and blinkenlighten just creep back in - as if the designers are scared that punters will see "less chrome" as "less power". My phone has about a dozen buttons scattered about its periphery PLUS a thumbwheel PLUS a touch screen/stylus PLUS a slide out keyboard - and while you can pretty much do anything using just buttons, just touch-screen or just keyboard, you need a lot of practices and cover-to-cover RTFMing. Usually, you end up using an inefficient combo of all of them (untilyou drop the stylus). Apple have the cojones to say "no - you're not having any buttons or a slide-out keyboard" and, if they put all their efforts into making the touchscreen work really intuitively, they could have a winner that will "grow" the market for powerful smartphones.
The non-3G thing seems "interesting" though - perhaps it makes sense in the USA but I assume that they don't plan to launch in Europe without 3G or better...
So does anybody have a magic solution that gives the client the right to use the original work that they've paid for, without the developer having to give up the rights to every last generic utility class they wrote, and consequently having to clean-room their next project, lest the original client sells out to a litigation-only troll company?
(Yes, but apart from that, Dr Stallman...)
The amount of dilligence that can reasonably be expected from a developer rather depends on the length and scale of the project - if its a 5 year project to write a new operating system that's one thing, but if its just a 6-week website job the client can hardly expect to pwn your browser-sniffing code...
I'll take your word for it, having never tried an NTSC conversion of a UK machine, but I'd point out that it ought to have been more than possible to fit the 176 vertical pixels of the spectrum's display within the 240 lines (200 in the safe area) of a single field of an NTSC frame.
OK, maybe the Speccie was not bitten by this one, but the BBC Micro (256 lines) certainly was.
The BBC *did* have 80 column text - MODE 0 and MODE 6.
I actually meant "a more usable 80 column mode" but didn't want to write a lecture on the BBC Micro.
Using those modes ate most of the RAM (20K out of 32K for MODE 0 - a bit less for MODE 6 at the expense of fugly black horizontal lines over coloured backgrounds or vertical lines) - and they were only "monochrome" (or 2 colour or whatever you want to call it). This was a huge problem for "business" software written in BASIC - slightly less if the software was machine code and blown into ROM. That's why I said extra memory also fixed the problem. Any self-respecing BBC nerd had long since lost the screws holding the lid on, and had at least 3 third-party piggy-back boards with dedicated video RAM, paged RAM for soft-loading "ROM" images, high-density disc controllers etc. but this could be a bit offputting for "serious" customers, but not all software supported them. The official "B+" model with these things properly integrated was rather belated.
The machine may well have done well in the US too, had Timex -- the company building the machine under license in the US -- wasn't already in financial trouble and about to fold.
There were 3 big barriers (at the time) to stop British machines taking off in the US:
The TV standard - NTSC has less scan lines than PAL, so while a US computer could easily be tweaked to output frequencies that a PAL TV could cope with, going the other way tended to mean losing a row or two of text or graphics from the screen - which broke any software with the screen size hard-coded in (which, in those days, was most of it).
EM emission standards. At the time, I don't think the UK had got round to regulating this and a Speccy or BBC Micro had no EM shielding and would wipe out any FM radio within earshot, so cases etc. had to be redesigned to accommodate EM shielding.
Price - despite the Pound varying between $1.40 and $2 over the years, there is a long and continuing (*cough* Microsoft *cough*) history of US firms setting UK prices by crossing out the $ sign to a £. In the UK, Spectrum vs. C64 argument tended to be a non-argument because the former was so cheap. This advantage tended to evaporate once the computers hit the US market. Once the price advantage was removed it became kinda obvious that the C64 was better built and had more sophisticated hardware than the Speccy. Likewise, the Apple ][ was fairly unaffordable in the UK and never had much market share - leaving an open niche for the BBC Micro that didn't exist in the US. Sadly, instead of consolidating this niche by producing a BBC with more memory and 80-column text (actually, the first would have enabled the latter) Acorn tried to compete with Sinclair by prodicing a BBC-with-all-the-good-features-removed and lost ground.
Apple: You may never, under any circumstances, on any hardware, at any time, for any reason, ever run OS X under virtualization. Period.
Slashdot's response: God I hate Microsoft!
Has Apple said "never, ever" to virtualization, or is it just that negotiating with Apple over how to do it legally is not on Parallels/VMWare's "TO DO" list (while they're busy racing to grab the lucrative windows-on-Mac market)?
Anway, if you don't like Apple's policy then it is incredibly easy to avoid buying a Mac because Apple do not have a 95%+ monopoly in the personal computer market - the only problem is which alternative you choose because Microsoft have a 95%+ monopoly in the PC market so even if you plump for Linux or BSD you'll find that lots of people take for granted that you can run Windows software.
A lot of good software is Windows only because, what with Microsoft having a 95%+ monopoly in the PC market its quite hard for software houses to justify supporting other platforms.
So, if a demand for virtualized Mac OSX does develop and Apple continue to block it then Apple will lose business. Microsoft, however, have a 95%+ monopoly in the PC market and can get away with all sorts of customer-hostile tricks - forbidding virtualization of the cheaper Vista versions doesn't impact on their income from the "Microsoft tax" on new computers and it doesn't really affect the big, corporate, volume licensing clients much. The people who it affects disproportionately are those using Macs and Linux who need to use a few Windows apps - not only do they (technically) have to fork out for a "full version" of Windows - already 2-3 times the retail price of the OEM version - they now have to buy the most expensive version too (or will do when XP is no longer easily available).
P.S. did I mention that Microsoft have a 95%+ monopoly in the PC market - which is why slashdot (plus the authoriities in every country that has any sort of monopoly/antitrust legislation) apply different standards to Microsoft and Apple.
Some day, either Apple will someday address the technical and legal issues of virtualizing OS X or they will face the commercial consequences - unlike Microsoft who have a 95%+ monopoly and can give commercial consequences the finger.
I'm sure that VMWare and/or Parallels would be delighted to work with Apple on implementing some safeguards to ensure that it will only run on VMs hosted on Mac hardware - if they see a demand. However, the main reason for wanting to virtualize OS X are:
Hackers wanting to run OS X on their PCs to show off. I doubt Apple will lose much sleep over this, provided that potential switchers are not put off by half-baked, flakey "demos" of OS X.
Server virtualization. How popular is OS X for the sort of web/application serving that benefits from virtualization, though? I assumed that the main uptake was for file serving and render farms in Mac shops.
Developers. Problem 1: the PC world has armies of in-house corporate developers creating a huge market for things like VMWare workstation. Problem 2: currently, a lot of "legacy" testing involves PowerPC machines;
The last two are the important markets but are "academic" until there is a stable, mature virtualization product for Mac. Personally, I'm happy with Parallels as a way of running Windows apps, trying out Linux distros etc. but, looking at the support forums, I wouldn't trust it for "production" yet. Both Parallels and VMWare are capable of delivering an industry strength product but are, clearly, currently concentrating on the big market, which is a consumer product for running Windows.
fully spec'd with dual 30" monitors and tons o' RAM/HD still over $10K... bummer)
The only real concern is the RAM and Graphics:
since the MacPro uses those expensive FB-DIMMS with custom super-strength heatsinks. At least Crucial et. al. do suitable RAM now, a bit cheaper than Apple, but not so cheap that you'd want to chuck the 1GB supplied. The latter is a pain if you're shooting for 8-16GB and/or worry about the complicated bits of advice about balancing RAM between channels.
Apart from that, go buy cheap(er) standad SATA drives and monitors.
The graphics card range seems sensible for "pro" use but a bit sucky for games purposes - but since that generally means "windows games" I can see why Apple isn't falling over itself to support this. Anyway, the ball is in Nvidia/ATI's courts to produce more EFI-compatible cards.
I agree, though, there is a hole in Apple's range the shape of a mini-tower with a single CPU socket (i.e. up to quad core) a couple of PCIExpress slots and space for an extra hard drive or two.
Applicaton software is the main drag anchor that ties us to the x86, and currently a lot of stuff is still C/C++ compiled down to bare-metal machine code.
I'd guess that long term, more and more software will run as bytecode for a virtual machine, or in a scripting language of some type. Look how much already works that way: Java, Flash, AJAX,.Net, "LAMP". Then you have code translators and emulators like Rosetta on the mac for legacy apps.
If these start to consolidate a bit (e.g. if Google takes over the world and everything goes AJAX) then, rather than a new platform needing a critical mass of applications to be ported individually to be viable, all you need to port would be a kernel and half-a-dozen virtual machines.
Of course, that doesn't overcome the massive "sticktion" of the x86 installed base.
In a sense, that's already happening - in hardware - as post-586 Intel chips are effectively a "RISC*" core with a (hardware) translator turning x86 code into core instructions.
* the term RISC is getting less applicable as more specialist bolt-ons for vector/multimedia work get added.
In many cases, especially for applications as opposed to system/low level software, yes, or with a modest amount of patching. If there is something that depends on byte sex or other processor-specific features then its quite likely that the source already contains conditional compilation to support several processors. Its not just a technical thing - portability between different architectures and UN*X flavours is part of the culture, whereas (apart from a brief interlude when NT was multiplatform) Windows has usually been tightly tied to, not just the x86, but the "IBM PC" implementation thereof.
If so, then why aren't there more distros that support PowerPC.
Because the vast majority of the demand is for x86. If you want a Linux box, the price difference between "commodity" PC hardware and specialist non-x86 hardware is usually end-of-argument. The only significant installed base of PPC desktop computers is probably pre-2006 Macs: given that the main reason for buying a Mac is to run Mac OS; that Mac users can get their UNIX fix from OSX and that quite a lot of the Linux/FOSS software base has been ported to OS X, there are so many distros that the PPC Mac can support.
If you look at the wider world of consoles, PDAs, routers/NAS then you'll find more ARM/PPC distros, but they're not generally the cuddly Ubutntu type and often start "warning: installing this incorrectly could turn your PDA/Playstation/Router into an expensive brick!".
A general switch to Linux would remove some of the roadblocks to a more hetrogenous world but it would not magically overcome the inertia.
Complete albums from EMI Music artists purchased on the iTunes Store will automatically be sold at the higher sound quality and DRM-free, with no change in the price.
Sounds more like it (decent artists ought to be able to put together 45 minutes of good music). Still need an argument as to why its better than ordering a CD and having to wait a whole day...
VPN access to home from work/mobile - "properly" through openvpn, or using ssh port forwarding (often easier).
Long downloads with ctorrent, wget etc.
I'm using it with an external USB hard-drive, but its quite possible to use a USB key (in which case it would be totally silent).
One hint - there are several alternative "firmware" packages that you can choose from: I went for "openSlug" which completely replaces the original firmware with a mini linux distro. I'd probably advise going with the superficially more kludgey-sounding "unslung" that keeps the original Linksys (linux-based) NAS system but lets you add packages. It seems to have a better range of packages and keeps the web interface.
If you want something that "just works" rather than a several days of cracking nerdy fun, an increasing number of network-attached hard drives have bittorrent clients, FTP servers etc. out-of-the-box e.g. this (Not a recommendation, just an example).
No, it doesn't use glibc. Think about it -- when the kernel boots up, it hasn't even mounted the root partition, how is it supposed to link in any libraries? And it's certainly not going to statically link glibc, it's far too big for that.
...and you'll find bits saying things like "hand optimized from GNU libc". Admittedly its only a "mountain of code" if you use SCO units (in which a real mountain must be the size of Jupiter), but its there.
I still don't understand why Tivo-ization is viewed as "bad" from people who want to develop or used GPL software.
Well, it certainly breaks the spirit. But the bogeyman is that (as someone else has posted elsewhere on the thread) TiVO is the tip of the iceberg and, sometime in the future the big evil closed software people will collude to ensure that new PCs will only run signed code - a bit like most games consoles.
That would be poison to free software as we know it, but big players could, potentially, pay the robber barrons to sign their GPL binaries for their customers and still "satisfy" GPLv2 by putting the - unsigned and useless - source on their websites.
But if that happened, it would probably because one or two powerful multinationals had enough market dominance and political influence to foist it on the industry - the only way the GPL can fight that is if it has a large user base including powerful friends in industry. Otherwise, all it can do is add a "suicide clause" that will ensure that free software becomes extinct the millisecond the "Securing Computing For Our Children" act is passed by RentAGov, instead of lingering on in captivity for a few years.
The danger is, if that clause misfires, excludes GPL software from legitimate cryptography and security applications, prevents signing firmware that has to meet legal standards (e.g. WiFi in many countries), complicates support and warranty for commercial GPL products etc. then it could damage the commercial uptake of FOSS and, when the Trusted Computing brigade tries it on, the big customers won't be there to shout "No!".
Now, the "GPLv3 means you have to give away your private keys" scare seems to have been debunked, and the latest GPL draft has clarified that section beyond recognition - probably why Linus is sounding happier.
However, I think these scares arose because the FSF were reading the GPL3 from their point of view as part of a select minority who have both legal AND technical knowledge, defending their own baby. OTOH, Linus and co seemed to be reading it from a more cynical "What's the marginally competent and hyper-defensive Legal department at your company going to tell the Linux-skeptic Boss (who's just read this article in PHBWeekly about how using free software will let terrorists rape your kids) that this means?".
Er, isn't it kinda derivative of the GNU standard C library? OK, that's LGPL we're talking about GPL here, but presumably the LGPLv3 will just be GPLv3 with suitable exceptions. Apart from that I've no idea whether any of the kernel is derivative of FSF "property" - the great thing about the GPL is that there's currently no just cause or impediment why it shouldn't be.
Anyway, with the possible exception of the "contacting the authors" issue, there's nothing kernel-specific about the criticisms of GPLv3.
No, vitally important news for the future of the free/open source software movement day.
The linux kernel is pretty important to (duh) most linux distributions. However, so is a load of Free Software Foundation-controlled stuff, not least the compilers, make tools, standard C libraries, and shedloads of userland utilities from the "ls" command through to EMACS... plus the GPL license itself. If the two factions fall out then it can only be bad for Linux and other FOSS.
Slighty satirized and only approximately true capsule summary of the problem:
The FSF wants - quite badly - to move to the GPLv3 to prevent "TiVOization" (using GPL code in a hardware device but with DRM-type tricks that stops users changing the code) and, more recently, to stop future Novell/Microsoft FUD campaigns.
Linus and other linux kernel contributors want - quite badly - to keep the GPLv2 because:
if it ain't broke, don't fix it;
TiVO etc. may be irksome but isn't worth the risk of "fixing" the GPLv2 (as programmers they understand this!)
Did we mention "If it aint broke, don't fix it"?
Previous drafts of the GPLv3 contained scary-sounding clauses about patents and use of encryption that, whatever their intention or precise legal meaning, would have had commercial GPL users running for the long grass.
Unlike FSF, "Linux" doesn't ask contributors to hand over copyright - so while FSF can change the license for the next version of gcc at the stroke of a pen, "Linux" can't change the license on the kernel without getting approval from hundreds of people, some of whom have inevitably emmigrated, died, gone to jail or, tragically, got jobs at Microsoft.
"Look, I was up burning the midday oil the other week because I decided to 'just fix' some code that wasn't really that broken so, take it from me, if it ain't seriously broke don't fix it!"
The pro-FSF lobby countered these concerns with:
Trust us, we're lawyers and academics
Feel free to comment on the detailed wording but we're not changing our mind about the principles
If you're against GPLv3 you must be for software patents and TiVOization
At which point ISTR Linus (or someone claiming to be he) said a Bad Word on Groklaw and PJ made him go and stand in the Naughty Corner until he had learned to control his potty mouth:-)
Then when the new draft of the GPLv3 appears it turns out that although the FSF have stuck to their guns they have been listening and have done some substantial re-drafting.
If Linus and the FSF are talking nicely again it can only be good news - even if Groklaw's swear box takings go down.
... and drastically reduce the load on the pagefile.
Seriously, the way of the future is surely to loose these distinctions between RAM and mass storage. Just leave it up to the OS to juggle data between CPU Cache, fast RAM, slow RAM, Flash and the internet, and refer to every bit of data by its IPv6 address...:-)
They're also mostly independent films.. which the article talks about.
...but then so are a lot of non-genre, critically acclaimed films.
TFA starts OK, pointing out the stigma against SF in any media, then goes off on a "blame Lucas" tangent. This is just nostalgia towards pre-Star Wars days: "Forbidden Planet" may be a great movie, but it was basically an (invisible) monster flick with lines like "raised almost literally to the power of infinity" and "any quantum mechanic in the service...". Likewise, "The Day The Earth Stood Still" bears less relation to the source material ("Farewell to the Master" by Harry Bates) than most of the P.K.Dick adaptations. The pre-Lucas versions "War of the Worlds" and "The Time Machine" were both "dumbed down" and actionized c.f. the books.
and AI was just a crappy remake of D.A.R.Y.L... without the heart.
Sorry. I liked AI. The film was pervaded by a sense of outrage that even if the kid was "just a machine" and couldn't be exploited, his adoptive mother certainly had been. It was nicely ambiguous as to whether the kid was feeling genuine emotions or was just obsessively following his programming. The androids were cleverly positioned in "uncanny valley" - almost, but not quite human, and not quite understanding the world around them, with even the superficially world-wise loverbot unable to distinguish fantasy from reality. Because of this, all the superficially schmaltzy stuff had a disturbing edge. OK, I'd recomend hitting stop before the epilogue - although even that can be interpreted differently.
You can add D.A.R.Y.L. to the list if you like - but if you do I'll start defending "I, Robot". (Just joking - the obtrusive product placement and portrayal of Susan Calvin as a hottie outweigh any possible defence!)
God, I hate Macromedia, and the marketing departments that are addicted to it.
Yeah, don't you just hate the way streamed Flash videos "just work" across PC and Mac, with IE, Firefox, Safari or Opera - for anybody with the ubiquitous Flash plug-in installed?
Of course, they discriminate against Linux users... Oh, wait, no, I just clicked "Install Plugin" in Firefox under Ubuntu and the video was running within 15 seconds... OK, so its not so easy if you're running PPC or 64 bit, and I had to click-through my Immortal Soul to Adobe, but they'd have to fight Microsoft, the Inland Revenue and T-Mobile if they wanted to collect that!
Yes, Flash is propietary, and yes, it has been abused for unnecessary eye candy, but it is also a bloody good product for small-scale, web-deliverable, cross-platform multimedia/forms apps that don't need the full might (and bewildering technology thicket) of Java. Show me (say) a combination of SVG and ECMAScript that actually works reliably.
PS "works reliably" in this context precludes telling your users/clients/potential customers "What? You're using Firefox|IE|Safari under Windows|MacOS? Just install Gentoo and do 'sudo apt-get-install mypersonalideaofadecentbrowserV1.0E-12PreAlpha' you moron! "
Slashdot Mirror
What? Next thing you'll be telling me is that my Kryptonite pendant won't actually disable any passing superhumans!
Does that mean that they are going to honour this request from the NeoOffice people?
Meanwhile...
The only way* for a company to make "huge amounts of revenue" from Open Source software is to add value so that people are prepared to pay you money for something that they could get elsewhere for free. That "value" might be providing top quality support, or it might be investigating in marketing or just having a number of employees who wear suits and use words like "leverage" that give corporate clients a warm fuzzy feeling. Either way, does anybody really have a problem with that?
Any company director who looses sleep about getting all this "money for nothing" simply needs to let their employees use some of their paid time to contribute to writing OSS code or coordinating OSS development.
*(excluding the "extort protection money on the back of questionable IP violation claims" method, of course).
But inside every pencil is an inanimate carbon rod and I'm sure we all know what a vital resource that can be in space travel.
PS: I am fully cogniscent that accounts of NASA's profligate spending on microgravity-enabled ballpoint pens are widely disputed and most probably false - however, the widespread recognition of the abovementioned meme facilitated the employment of a comedic device known as a "double entendre" in which apparently innocent phrases quoted in a suitable context become amenable to an alternative interpretation of a sexual, or otherwise taboo, nature, to humourous effect. Also, for the further avoidance of doubt, I do not have reliable evidence for the insinuation that Russian cosmonauts make improper use of writing implements or, indeed, that said implements can be satisfactorally used for auto-erotic purposes. Finally, I believe that the assertion that all Slashdot readers are celebate, if not virgin, is unjustified and that many have, in fact, experienced sexual relations - sometimes even with other people.
I apologise profusely for neglecting to negate whatever modest humour my original posting may have possessed by introducing the above caveats at an earlier stage in proceedings.
That is, of course, a valid solution, but it might not suit everybody. :-)
A: Spend $100,000,000 developing high tech, er, appliances that work in zero gravity, then brace for the ensuing scandal when it emerges that the Russians just used pencils...
Alternatively, recruit more nerds and less jocks. Why not advertise on Slashdot?
...but iPhone and Windows Mobile aren't competing with the 1.3 billion phones sold. They are competing for the same, relatively small subset of the market representing high-end smartphones with EMAIL and serious web-browsing facilities.
Now, in the MS corner, we have Windows Mobile. I have a fairly high-end Windows smartphone, and while I like the pwer of the thing, using it can be like kicking a dead whale along a beach. Using it as a music player is particularly excruciating. I would be very reluctant to recommend it to a non techie. Mind you, some basic phones are pretty nasty to use, as well.
Although nobody has seen an iPhone properly, Apple have a track record and would need to be having a very bad day not to produce something vastly more usable. Its not like XP vs. OSX: personally, ignoring security and reliability, I don't think that there is any clear blue water between OSX and Win2K/XP on the usability front - I actually prefer the XP GUI in some ways - but Windows Mobile feels like a throwback to Windows 2.
Its not that Apple stuff is perfect - just that it usually gives the impression that it was actually designed by people who gave a damn about the product in a world where many - if not most - other computer and home electronics products seem to have congealed out of a committee process.
Also - Apple seem to be capable of making "less is more" decisions. Notice that hardly anybody has matched Apple's minimalist design style? Other manufacturers have produced designer-y ranges but the extra buttons, chrome grills, go-faster stripes and blinkenlighten just creep back in - as if the designers are scared that punters will see "less chrome" as "less power". My phone has about a dozen buttons scattered about its periphery PLUS a thumbwheel PLUS a touch screen/stylus PLUS a slide out keyboard - and while you can pretty much do anything using just buttons, just touch-screen or just keyboard, you need a lot of practices and cover-to-cover RTFMing. Usually, you end up using an inefficient combo of all of them (untilyou drop the stylus). Apple have the cojones to say "no - you're not having any buttons or a slide-out keyboard" and, if they put all their efforts into making the touchscreen work really intuitively, they could have a winner that will "grow" the market for powerful smartphones.
The non-3G thing seems "interesting" though - perhaps it makes sense in the USA but I assume that they don't plan to launch in Europe without 3G or better...
So does anybody have a magic solution that gives the client the right to use the original work that they've paid for, without the developer having to give up the rights to every last generic utility class they wrote, and consequently having to clean-room their next project, lest the original client sells out to a litigation-only troll company?
(Yes, but apart from that, Dr Stallman...)
The amount of dilligence that can reasonably be expected from a developer rather depends on the length and scale of the project - if its a 5 year project to write a new operating system that's one thing, but if its just a 6-week website job the client can hardly expect to pwn your browser-sniffing code...
OK, maybe the Speccie was not bitten by this one, but the BBC Micro (256 lines) certainly was.
I actually meant "a more usable 80 column mode" but didn't want to write a lecture on the BBC Micro.
Using those modes ate most of the RAM (20K out of 32K for MODE 0 - a bit less for MODE 6 at the expense of fugly black horizontal lines over coloured backgrounds or vertical lines) - and they were only "monochrome" (or 2 colour or whatever you want to call it). This was a huge problem for "business" software written in BASIC - slightly less if the software was machine code and blown into ROM. That's why I said extra memory also fixed the problem. Any self-respecing BBC nerd had long since lost the screws holding the lid on, and had at least 3 third-party piggy-back boards with dedicated video RAM, paged RAM for soft-loading "ROM" images, high-density disc controllers etc. but this could be a bit offputting for "serious" customers, but not all software supported them. The official "B+" model with these things properly integrated was rather belated.
There were 3 big barriers (at the time) to stop British machines taking off in the US:
Has Apple said "never, ever" to virtualization, or is it just that negotiating with Apple over how to do it legally is not on Parallels/VMWare's "TO DO" list (while they're busy racing to grab the lucrative windows-on-Mac market)?
Anway, if you don't like Apple's policy then it is incredibly easy to avoid buying a Mac because Apple do not have a 95%+ monopoly in the personal computer market - the only problem is which alternative you choose because Microsoft have a 95%+ monopoly in the PC market so even if you plump for Linux or BSD you'll find that lots of people take for granted that you can run Windows software.
A lot of good software is Windows only because, what with Microsoft having a 95%+ monopoly in the PC market its quite hard for software houses to justify supporting other platforms.
So, if a demand for virtualized Mac OSX does develop and Apple continue to block it then Apple will lose business. Microsoft, however, have a 95%+ monopoly in the PC market and can get away with all sorts of customer-hostile tricks - forbidding virtualization of the cheaper Vista versions doesn't impact on their income from the "Microsoft tax" on new computers and it doesn't really affect the big, corporate, volume licensing clients much. The people who it affects disproportionately are those using Macs and Linux who need to use a few Windows apps - not only do they (technically) have to fork out for a "full version" of Windows - already 2-3 times the retail price of the OEM version - they now have to buy the most expensive version too (or will do when XP is no longer easily available).
P.S. did I mention that Microsoft have a 95%+ monopoly in the PC market - which is why slashdot (plus the authoriities in every country that has any sort of monopoly/antitrust legislation) apply different standards to Microsoft and Apple.
Some day, either Apple will someday address the technical and legal issues of virtualizing OS X or they will face the commercial consequences - unlike Microsoft who have a 95%+ monopoly and can give commercial consequences the finger.
I'm sure that VMWare and/or Parallels would be delighted to work with Apple on implementing some safeguards to ensure that it will only run on VMs hosted on Mac hardware - if they see a demand. However, the main reason for wanting to virtualize OS X are:
The last two are the important markets but are "academic" until there is a stable, mature virtualization product for Mac. Personally, I'm happy with Parallels as a way of running Windows apps, trying out Linux distros etc. but, looking at the support forums, I wouldn't trust it for "production" yet. Both Parallels and VMWare are capable of delivering an industry strength product but are, clearly, currently concentrating on the big market, which is a consumer product for running Windows.
The only real concern is the RAM and Graphics:
since the MacPro uses those expensive FB-DIMMS with custom super-strength heatsinks. At least Crucial et. al. do suitable RAM now, a bit cheaper than Apple, but not so cheap that you'd want to chuck the 1GB supplied. The latter is a pain if you're shooting for 8-16GB and/or worry about the complicated bits of advice about balancing RAM between channels.
Apart from that, go buy cheap(er) standad SATA drives and monitors.
The graphics card range seems sensible for "pro" use but a bit sucky for games purposes - but since that generally means "windows games" I can see why Apple isn't falling over itself to support this. Anyway, the ball is in Nvidia/ATI's courts to produce more EFI-compatible cards.
I agree, though, there is a hole in Apple's range the shape of a mini-tower with a single CPU socket (i.e. up to quad core) a couple of PCIExpress slots and space for an extra hard drive or two.
Applicaton software is the main drag anchor that ties us to the x86, and currently a lot of stuff is still C/C++ compiled down to bare-metal machine code.
I'd guess that long term, more and more software will run as bytecode for a virtual machine, or in a scripting language of some type. Look how much already works that way: Java, Flash, AJAX, .Net, "LAMP". Then you have code translators and emulators like Rosetta on the mac for legacy apps.
If these start to consolidate a bit (e.g. if Google takes over the world and everything goes AJAX) then, rather than a new platform needing a critical mass of applications to be ported individually to be viable, all you need to port would be a kernel and half-a-dozen virtual machines.
Of course, that doesn't overcome the massive "sticktion" of the x86 installed base.
In a sense, that's already happening - in hardware - as post-586 Intel chips are effectively a "RISC*" core with a (hardware) translator turning x86 code into core instructions.
* the term RISC is getting less applicable as more specialist bolt-ons for vector/multimedia work get added.
In many cases, especially for applications as opposed to system/low level software, yes, or with a modest amount of patching. If there is something that depends on byte sex or other processor-specific features then its quite likely that the source already contains conditional compilation to support several processors. Its not just a technical thing - portability between different architectures and UN*X flavours is part of the culture, whereas (apart from a brief interlude when NT was multiplatform) Windows has usually been tightly tied to, not just the x86, but the "IBM PC" implementation thereof.
Because the vast majority of the demand is for x86. If you want a Linux box, the price difference between "commodity" PC hardware and specialist non-x86 hardware is usually end-of-argument. The only significant installed base of PPC desktop computers is probably pre-2006 Macs: given that the main reason for buying a Mac is to run Mac OS; that Mac users can get their UNIX fix from OSX and that quite a lot of the Linux/FOSS software base has been ported to OS X, there are so many distros that the PPC Mac can support.
If you look at the wider world of consoles, PDAs, routers/NAS then you'll find more ARM/PPC distros, but they're not generally the cuddly Ubutntu type and often start "warning: installing this incorrectly could turn your PDA/Playstation/Router into an expensive brick!".
A general switch to Linux would remove some of the roadblocks to a more hetrogenous world but it would not magically overcome the inertia.
Sounds more like it (decent artists ought to be able to put together 45 minutes of good music). Still need an argument as to why its better than ordering a CD and having to wait a whole day...
That's the original combined Perl/Python scripting language, BTW, not the Perl 6 virtual machine that ripped off the name.
I use a "hacked" Linksys NSLU2 to run:
I'm using it with an external USB hard-drive, but its quite possible to use a USB key (in which case it would be totally silent).
One hint - there are several alternative "firmware" packages that you can choose from: I went for "openSlug" which completely replaces the original firmware with a mini linux distro. I'd probably advise going with the superficially more kludgey-sounding "unslung" that keeps the original Linksys (linux-based) NAS system but lets you add packages. It seems to have a better range of packages and keeps the web interface.
If you want something that "just works" rather than a several days of cracking nerdy fun, an increasing number of network-attached hard drives have bittorrent clients, FTP servers etc. out-of-the-box e.g. this (Not a recommendation, just an example).
True, but try: /usr/src/linux
.* free software foundation" .
cd
grep -ri "copyright
...and you'll find bits saying things like "hand optimized from GNU libc". Admittedly its only a "mountain of code" if you use SCO units (in which a real mountain must be the size of Jupiter), but its there.
Well, it certainly breaks the spirit. But the bogeyman is that (as someone else has posted elsewhere on the thread) TiVO is the tip of the iceberg and, sometime in the future the big evil closed software people will collude to ensure that new PCs will only run signed code - a bit like most games consoles.
That would be poison to free software as we know it, but big players could, potentially, pay the robber barrons to sign their GPL binaries for their customers and still "satisfy" GPLv2 by putting the - unsigned and useless - source on their websites.
But if that happened, it would probably because one or two powerful multinationals had enough market dominance and political influence to foist it on the industry - the only way the GPL can fight that is if it has a large user base including powerful friends in industry. Otherwise, all it can do is add a "suicide clause" that will ensure that free software becomes extinct the millisecond the "Securing Computing For Our Children" act is passed by RentAGov, instead of lingering on in captivity for a few years.
The danger is, if that clause misfires, excludes GPL software from legitimate cryptography and security applications, prevents signing firmware that has to meet legal standards (e.g. WiFi in many countries), complicates support and warranty for commercial GPL products etc. then it could damage the commercial uptake of FOSS and, when the Trusted Computing brigade tries it on, the big customers won't be there to shout "No!".
Now, the "GPLv3 means you have to give away your private keys" scare seems to have been debunked, and the latest GPL draft has clarified that section beyond recognition - probably why Linus is sounding happier.
However, I think these scares arose because the FSF were reading the GPL3 from their point of view as part of a select minority who have both legal AND technical knowledge, defending their own baby. OTOH, Linus and co seemed to be reading it from a more cynical "What's the marginally competent and hyper-defensive Legal department at your company going to tell the Linux-skeptic Boss (who's just read this article in PHBWeekly about how using free software will let terrorists rape your kids) that this means?".
Er, isn't it kinda derivative of the GNU standard C library? OK, that's LGPL we're talking about GPL here, but presumably the LGPLv3 will just be GPLv3 with suitable exceptions. Apart from that I've no idea whether any of the kernel is derivative of FSF "property" - the great thing about the GPL is that there's currently no just cause or impediment why it shouldn't be.
Anyway, with the possible exception of the "contacting the authors" issue, there's nothing kernel-specific about the criticisms of GPLv3.
No, vitally important news for the future of the free/open source software movement day.
The linux kernel is pretty important to (duh) most linux distributions. However, so is a load of Free Software Foundation-controlled stuff, not least the compilers, make tools, standard C libraries, and shedloads of userland utilities from the "ls" command through to EMACS... plus the GPL license itself. If the two factions fall out then it can only be bad for Linux and other FOSS.
Slighty satirized and only approximately true capsule summary of the problem:
The FSF wants - quite badly - to move to the GPLv3 to prevent "TiVOization" (using GPL code in a hardware device but with DRM-type tricks that stops users changing the code) and, more recently, to stop future Novell/Microsoft FUD campaigns.
Linus and other linux kernel contributors want - quite badly - to keep the GPLv2 because:
The pro-FSF lobby countered these concerns with:
At which point ISTR Linus (or someone claiming to be he) said a Bad Word on Groklaw and PJ made him go and stand in the Naughty Corner until he had learned to control his potty mouth :-)
Then when the new draft of the GPLv3 appears it turns out that although the FSF have stuck to their guns they have been listening and have done some substantial re-drafting.
If Linus and the FSF are talking nicely again it can only be good news - even if Groklaw's swear box takings go down.
... and drastically reduce the load on the pagefile.
Seriously, the way of the future is surely to loose these distinctions between RAM and mass storage. Just leave it up to the OS to juggle data between CPU Cache, fast RAM, slow RAM, Flash and the internet, and refer to every bit of data by its IPv6 address... :-)
...but then so are a lot of non-genre, critically acclaimed films.
TFA starts OK, pointing out the stigma against SF in any media, then goes off on a "blame Lucas" tangent. This is just nostalgia towards pre-Star Wars days: "Forbidden Planet" may be a great movie, but it was basically an (invisible) monster flick with lines like "raised almost literally to the power of infinity" and "any quantum mechanic in the service...". Likewise, "The Day The Earth Stood Still" bears less relation to the source material ("Farewell to the Master" by Harry Bates) than most of the P.K.Dick adaptations. The pre-Lucas versions "War of the Worlds" and "The Time Machine" were both "dumbed down" and actionized c.f. the books.
Sorry. I liked AI. The film was pervaded by a sense of outrage that even if the kid was "just a machine" and couldn't be exploited, his adoptive mother certainly had been. It was nicely ambiguous as to whether the kid was feeling genuine emotions or was just obsessively following his programming. The androids were cleverly positioned in "uncanny valley" - almost, but not quite human, and not quite understanding the world around them, with even the superficially world-wise loverbot unable to distinguish fantasy from reality. Because of this, all the superficially schmaltzy stuff had a disturbing edge. OK, I'd recomend hitting stop before the epilogue - although even that can be interpreted differently.
You can add D.A.R.Y.L. to the list if you like - but if you do I'll start defending "I, Robot". (Just joking - the obtrusive product placement and portrayal of Susan Calvin as a hottie outweigh any possible defence!)
Yeah, don't you just hate the way streamed Flash videos "just work" across PC and Mac, with IE, Firefox, Safari or Opera - for anybody with the ubiquitous Flash plug-in installed?
Of course, they discriminate against Linux users... Oh, wait, no, I just clicked "Install Plugin" in Firefox under Ubuntu and the video was running within 15 seconds... OK, so its not so easy if you're running PPC or 64 bit, and I had to click-through my Immortal Soul to Adobe, but they'd have to fight Microsoft, the Inland Revenue and T-Mobile if they wanted to collect that!
Yes, Flash is propietary, and yes, it has been abused for unnecessary eye candy, but it is also a bloody good product for small-scale, web-deliverable, cross-platform multimedia/forms apps that don't need the full might (and bewildering technology thicket) of Java. Show me (say) a combination of SVG and ECMAScript that actually works reliably.
PS "works reliably" in this context precludes telling your users/clients/potential customers "What? You're using Firefox|IE|Safari under Windows|MacOS? Just install Gentoo and do 'sudo apt-get-install mypersonalideaofadecentbrowserV1.0E-12PreAlpha' you moron! "