I fully expect that Sony will be putting out DVD / UMD combo players soon enough, but using some kind of Heath Robinson device for capturing your screen via a series of mirrors, lenses and a camera, is quite frankly, fucking stupid by a long margin.
It would be nice to see a PSP with some kind of expansion port, but this kind of thing really is ridiculous. Chances are the quality is on par with your average bootleg cinema-to-camcorder movie.
All the telcos have also bought into the same busted logic Apple - that people are going to pay ludicrous sums of money to watch what they can watch on their TV, or with a PVR, or with a computer for nothing at all. In the case of 3G phones, it's even worse since they're charging to watch movie trailers, and 5 minute clips.
I could possibly see such a service working, if it acted like a subscription based series of "channels". i.e. you paid $15 a month or whatever to watch a MTV-like channel, or a movie channel, or a comedy channel etc. that you primed with your favourite shows. I cannot see it working at all if they expect people to pay per show and to sit around the several hours for the show to actually download.
If SCO want to throw some money at MySQL for commercial support, then so what?. It might hasten SCO's demise, and the money can be used for bug fixing instead of lining some lawyer's pockets.
OpenOffice is LGPL'd and makes use of Mozilla, Java, Python and no doubt a large swathe of other libs and utilities. I don't see how the licence has been an impediment thus far.
I'd be more concerned that if it were GPL'd that it couldn't use some or all of the above. Now arguably, OO does need to shed some pounds so if it dumped Python and / or Java that might be no bad thing, but that's a different topic altogether.
I haven't tried.NET 2.0, but one of the few good things you can say about VS.NET 2003/2002 is that it is incredibly easy and fast to knock together a form. Trying to do the same for Java can be an exercise in frustration. On the downside,.NET doesn't have (until 2.0) layout models, so resizing must be coded manually and there are probably issues with theme engines / cross platform too.
Even after 5 years of Java development, I haven't seen a visual designer which I would describe as any better than tolerable. Currently I'm using VE in Eclipse. For the most part it works okay but sometimes it just destroys your form making it essential to maintain backups. On top of that, I don't like that when you you switch from one layout model to another that it royally screws up the constraints it uses.
How is that not true of any other language or platform in existence? The problem with those other choices you mention is that they introduce platform dependencies, rendering Java non-cross platform and essentially eliminating any reason to use it all. If Swing were an acceptable speed this would not be an issue, but performance is really, really poor.
I know there all kinds of rationale for cross-platform UIs, but Sun really, really need to fix Swing or deprecate it for SWT or something else. The thing is a frigging albatross when developing desktop Java apps.
C# or rather the CLR, is of comparable speed to Java. I use both day to day and don't perceive any difference between the two EXCEPT as far as UI performance goes. Windows.Forms is a lot faster than Swing, though probably no faster than SWT. This is due to it using native controls, and having no layout models to speak of. In fact Swing probably has no small part to play in the perceived speed of Java.
Startup of Java seems a little slower than CLR but not markedly slower. Mono's startup is hardly a speed demon either. In all, I think most of the hype around the CLR is just that - hype. The CLR is a lot nicer virtual machine, and C# is as close to C++ / Java that I couldn't care which language I'm writing to, but the performance differences are IMHO neglible.
As an example of how fast Java is, I have written a poker simulator which plays 1000 games from a deck of cards which is shuffled 1000 times for each game. The hands from each game are then sorted, and scored. It's very Vector and LinkedList intensive and each loop probably sees a dozen or so temporary Vectors or arrays used during comparison - i.e. 1,000,000 iterations with many more nested within. It takes just over a second on my 1.8 Ghz PC.
As far as I'm concerned, that's plenty fast. I reckon I could probably halve the time if I bothered to replace all the vectors with fixed arrays.
It would be hard to compare Delphi (as in traditional Delphi) to either C# or Java since it is a compiled language. I don't know what Borland are doing with it these days so I wouldn't be surprised if it's gone the CLR route too.
Although it doesn't have anywhere near the rich number of libraries C/C++ works with (I'd like to see someone make a cutting edge game in Java).
I disagree with the first part of that statement. Java hands down wipes the floor with C++ even if you deck it out with STL and Boost++. Not only does it have libraries for all the usual things language runtimes provide, but encryption, UI, database, graphics, sound plus a raft of enterprise level libs.
You might be cobble together libs for C++ which do these things, but I bet they don't run cross-platform. Even the likes of OpenGL require some platform specific code.
I don't believe you will ever see a cutting edge game in Java (unless you consider 2D orthogonal Puzzle Pirates to be cutting edge), but I bet you a hell of a lot of MMPORG type games are using Java somewhere in the backend. In fact if I were doing an MMPORG I'd probably code up most of the back-end stuff in Java unless we were talking about a game which *really* hammered the server. Most games would have components for account maintenance, maps, NPCs, object databases, RNGs etc. I don't think it would be any trouble to run substantial chunks of that on Java.
In my experience Java is fast enough for most server side operations and just as fast as the highly touted CLR / C#. The problem with Java is that Swing performance truly is wretched for writing UIs, but with SWT it's quite acceptable for most tasks.
The problem aside from Swing of using Java for UI work is the tools are wretched. Even something like the VE extension for Eclipse can't overcome the sheer bloody awfulness of trying to develop visual applications in the Java language.
Ender's Game doesn't need a great screenplay. There are no plot threads, no intricate dialogue to distill, no nothing that couldn't be easily transplanted from the book to the script. The challenge if there is any, would be add tension and excitement to what is a rather boring and repetitive book.
The ending is a zinger, but most of the stuff that precedes it, especially the dialog is rather average. In fact the dialogue is extremely clunky. I hear people say "well these kids are geniuses so they'd speak like adults", but that is more like someone making excuses for the poor dialogue. Kids, even precocious kids don't talk like the ones in Ender's Game.
a) A regular backup of personal and work related stuff to DVD+RW. Stored in a firesafe. b) A "personal" kit backup on USB 512Mb flash that I take on travel. This consists of my insurance photos, password safe, finance records, GPG keys, bookmarks, important docs and some software such as firefox, TomsRTBT Linux etc. c) An "emergency kit" encrypted self-extracting executable uploaded to a website. Just contains a mini version of insurance photos, password safe, finance records etc.
I hope between the three of them that I'm left with *something* if my house burnt to the ground or something else happened. The trouble is keeping them all in sync and it normally requires me to remember to manually copy files from here or there before going away somewhere. I'd probably write a script to automate the syncing if I could be bothered. I've been meaning to play around with Ruby or Python so maybe that would be an opportunity to do all this.
On top of that I've just gotten Subversion running on Win32. This just runs locally so it's not necessarily a backup in itself, but I've started to use it for storing documents and all my programming work. I also back it up to DVD+RW from time to time. I never bothered with source control before now when it was my own stuff, but since trying to use the seriously screwed visual editor in Eclipse, I'm starting to appreciate that a source level undo is actually useful.
And it's extremely rare that you would know the correct path unless you've thought carefully about where you intend going. And when you're leading a team if would be a good idea to let them know where you're going too.
Hence the reason that a spec, or at least a document covering the major points is essential to most projects. Even if you never read the thing again, the act of writing it ensures you've considered the major obstacles and choices that are necessary during the implementation.
Now perhaps specs and designs don't matter that much to Linus, but to be frank, he's walking a well worn path which is already heavily signposted, and he has many scouts ready who can wander off the path if need be and get lost for days and weeks. Most projects don't work like that and for good reason.
It's not "simpler" not "equally effective" to have "runners" getting memory cards. Supposedly a wi-fi camera has the option to upload immediately each foto after taken.
Yes it is. Much simpler. No wifi is required, no hotspots, no interference, no mobile phone - just a guy with a pair of legs. You assume this camera supports ad hoc uploading or "guerilla" style photography when in fact it would do no such thing. It is likely that everytime you wanted to upload you would have to flip the camera into a special wireless mode, wait for it to find a hotspot, go through the various "wizard" steps to authenticate, select your pics and then wait while they uploaded to the Kodak site. All while dodging the rubber bullets, water cannons and baton charges.
It would be faster and more effective to hand the card over to someone else who retreats from the front lines while you take another set of pics. Alternatively if you absolutely had to broadcast pics yourself, you'd be better off to do it from a pocket pc and software that dumps the entire contents of an inserted card to the net.
Except of course now you're expecting there to be wifi spots at the same places there are riots and civil disturbances. It's hard enough to find wifi access at the best of times, let alone in a pitched battle rolling back and forth between streets. And if there were, no doubt you'd have to stand quite still while your pics were uploaded which wouldn't necessarily be convenient at the time. If that weren't unlikely enough a totalitarian state is likely to have little internet access or extremely restricted access. On top of that is Kodak itself. Their site probably pitches itself as "family friendly" so you can bet that any civil disobediance pics would be wiped off their site without a second's thought.
I wouldn't diss the idea completely - after all if your camera would connect to an ad-hoc network you could perhaps arrange for someone with a PDA or small laptop to shadow you at some distance and broadcast the pics back to them, but it would still be an awkward arrangement. And its doubtful that this camera would help you do that.
Perhaps it's simpler and equally effective to use redundancy - multiple photographers, with each passing their filled memory cards to runners.
MS Office users (and users of other applications) have been able to print straight to PDF for years now. They had to install a printer driver which comes with Adobe Acrobat, but that's about it.
It makes more sense to do it this way than implement PDF into every app since a printer driver works with anything. For example I print receipts out from Firefox using the PDF printer.
Besides, who says they will use OpenOffice at all? They're just saying what data format they want to use, and not what tools they use to read or write it. Obviously OpenOffice is the frontrunner, but it needn't be the only one.
For example, if it costs $5 million as speculated by some official to move to OpenOffice and $50 million to update / retrain for MS Office 12, how much would it cost to stay put with the existing MS Office and spend money developing ODF filters for it? It must be doable, in fact it is doable. Microsoft have implied as much when they said they would develop read / write filters for their XML format in versions of Office going back to 2000. If they can do their format, then ODF must be a shoe-in too.
I'd rather see that happen (assuming they released the code afterwards) than see them go to OpenOffice. A filter for MS Word & Excel would be immeasurably more valuable than a couple of thousand OO seats.
I know about this, but the exchange plugin needs to be part of it too. It doesn't make any difference for home use but a calendar / email app which doesn't talk to Exchange is next to useless in a lot of corporate environments. I am actually surprised that someone with deep pockets like IBM hasn't thrown a ton of cash as bounty for whoever integrates the exchange plugin into Thunderbird.
It seems weird that no one has ported it before now. Outlook is a piece of shit. Really. I have to use Outlook 2002 day in day out and the only thing going for it is that it's an email and calendar built into one. The mail portion is on par with Netscape Communicator 4.x (actually worse since NS didn't lock up randomly while syncing to the server), the calendar is a simple PIM.
The only reason it even exists as far as I can make out is because once a company gets MS Exchange, it's about the only client that is designed to work with it.
Personally I'd love to see Evolution on Windows. Its presence might shake the tree a bit. Even better (for me) would be to see Sunbird / Thunderbird merged and using the Novell Exchange plugin. I think Thunderbird is a killer email app, but the lack of Exchange support hurts it in the business environment.
Upload/Download to/from eDonkey has been less than 5% for years.
I not really familiar with the eDonkey network, was eDonkey packed with spyware? It would explain a lot if it was. I know Kazaa was too which prompted someone to produce a Kazaa Lite with all the spyware, ads, cookies, trackers, affiliate links and everything else ripped out.
I thought the open source and decentralized eMule was the tool of choice for the eDonkey network, with Shareaza and other tools following closely behind.
2) There is already something for GNOME/KDE integration: a GTK theme engine based on Qt. Thus, GTK apps look like Qt/KDE ones. Of course, its only useful if you use KDE...
That's better than nothing, but really the API to the theme engine should be a simple set of C APIs that can be implemented very easily and loaded against GTK and QT. This API and its headers would be specified by freedesktop and authors just have to implement their themes against the API. When GTK / QT loads the engine, it enumerates all themes, loads the user's selected engine and calls the functions to draw buttons, get metrics, get system colours and everything else that the theme should do.
It shouldn't be necessary for GTK to piggyback over QT or vice versa - just define a nice spec which allows both widget sets to call a well defined API.
Slashdot Mirror
$20? Think yourself lucky. In the Ireland UMD discs cost 24. Who is actually buying them is a complete mystery though.
Not only horrible but a totally stupid idea.
I fully expect that Sony will be putting out DVD / UMD combo players soon enough, but using some kind of Heath Robinson device for capturing your screen via a series of mirrors, lenses and a camera, is quite frankly, fucking stupid by a long margin.
It would be nice to see a PSP with some kind of expansion port, but this kind of thing really is ridiculous. Chances are the quality is on par with your average bootleg cinema-to-camcorder movie.
I could possibly see such a service working, if it acted like a subscription based series of "channels". i.e. you paid $15 a month or whatever to watch a MTV-like channel, or a movie channel, or a comedy channel etc. that you primed with your favourite shows. I cannot see it working at all if they expect people to pay per show and to sit around the several hours for the show to actually download.
If SCO want to throw some money at MySQL for commercial support, then so what?. It might hasten SCO's demise, and the money can be used for bug fixing instead of lining some lawyer's pockets.
I'd be more concerned that if it were GPL'd that it couldn't use some or all of the above. Now arguably, OO does need to shed some pounds so if it dumped Python and / or Java that might be no bad thing, but that's a different topic altogether.
Even after 5 years of Java development, I haven't seen a visual designer which I would describe as any better than tolerable. Currently I'm using VE in Eclipse. For the most part it works okay but sometimes it just destroys your form making it essential to maintain backups. On top of that, I don't like that when you you switch from one layout model to another that it royally screws up the constraints it uses.
I know there all kinds of rationale for cross-platform UIs, but Sun really, really need to fix Swing or deprecate it for SWT or something else. The thing is a frigging albatross when developing desktop Java apps.
Startup of Java seems a little slower than CLR but not markedly slower. Mono's startup is hardly a speed demon either. In all, I think most of the hype around the CLR is just that - hype. The CLR is a lot nicer virtual machine, and C# is as close to C++ / Java that I couldn't care which language I'm writing to, but the performance differences are IMHO neglible.
As an example of how fast Java is, I have written a poker simulator which plays 1000 games from a deck of cards which is shuffled 1000 times for each game. The hands from each game are then sorted, and scored. It's very Vector and LinkedList intensive and each loop probably sees a dozen or so temporary Vectors or arrays used during comparison - i.e. 1,000,000 iterations with many more nested within. It takes just over a second on my 1.8 Ghz PC.
As far as I'm concerned, that's plenty fast. I reckon I could probably halve the time if I bothered to replace all the vectors with fixed arrays.
It would be hard to compare Delphi (as in traditional Delphi) to either C# or Java since it is a compiled language. I don't know what Borland are doing with it these days so I wouldn't be surprised if it's gone the CLR route too.
I disagree with the first part of that statement. Java hands down wipes the floor with C++ even if you deck it out with STL and Boost++. Not only does it have libraries for all the usual things language runtimes provide, but encryption, UI, database, graphics, sound plus a raft of enterprise level libs.
You might be cobble together libs for C++ which do these things, but I bet they don't run cross-platform. Even the likes of OpenGL require some platform specific code.
I don't believe you will ever see a cutting edge game in Java (unless you consider 2D orthogonal Puzzle Pirates to be cutting edge), but I bet you a hell of a lot of MMPORG type games are using Java somewhere in the backend. In fact if I were doing an MMPORG I'd probably code up most of the back-end stuff in Java unless we were talking about a game which *really* hammered the server. Most games would have components for account maintenance, maps, NPCs, object databases, RNGs etc. I don't think it would be any trouble to run substantial chunks of that on Java.
The problem aside from Swing of using Java for UI work is the tools are wretched. Even something like the VE extension for Eclipse can't overcome the sheer bloody awfulness of trying to develop visual applications in the Java language.
Ender's Game doesn't need a great screenplay. There are no plot threads, no intricate dialogue to distill, no nothing that couldn't be easily transplanted from the book to the script. The challenge if there is any, would be add tension and excitement to what is a rather boring and repetitive book.
The ending is a zinger, but most of the stuff that precedes it, especially the dialog is rather average. In fact the dialogue is extremely clunky. I hear people say "well these kids are geniuses so they'd speak like adults", but that is more like someone making excuses for the poor dialogue. Kids, even precocious kids don't talk like the ones in Ender's Game.
a) A regular backup of personal and work related stuff to DVD+RW. Stored in a firesafe.
b) A "personal" kit backup on USB 512Mb flash that I take on travel. This consists of my insurance photos, password safe, finance records, GPG keys, bookmarks, important docs and some software such as firefox, TomsRTBT Linux etc.
c) An "emergency kit" encrypted self-extracting executable uploaded to a website. Just contains a mini version of insurance photos, password safe, finance records etc.
I hope between the three of them that I'm left with *something* if my house burnt to the ground or something else happened. The trouble is keeping them all in sync and it normally requires me to remember to manually copy files from here or there before going away somewhere. I'd probably write a script to automate the syncing if I could be bothered. I've been meaning to play around with Ruby or Python so maybe that would be an opportunity to do all this.
On top of that I've just gotten Subversion running on Win32. This just runs locally so it's not necessarily a backup in itself, but I've started to use it for storing documents and all my programming work. I also back it up to DVD+RW from time to time. I never bothered with source control before now when it was my own stuff, but since trying to use the seriously screwed visual editor in Eclipse, I'm starting to appreciate that a source level undo is actually useful.
Hence the reason that a spec, or at least a document covering the major points is essential to most projects. Even if you never read the thing again, the act of writing it ensures you've considered the major obstacles and choices that are necessary during the implementation.
Now perhaps specs and designs don't matter that much to Linus, but to be frank, he's walking a well worn path which is already heavily signposted, and he has many scouts ready who can wander off the path if need be and get lost for days and weeks. Most projects don't work like that and for good reason.
Yes it is. Much simpler. No wifi is required, no hotspots, no interference, no mobile phone - just a guy with a pair of legs. You assume this camera supports ad hoc uploading or "guerilla" style photography when in fact it would do no such thing. It is likely that everytime you wanted to upload you would have to flip the camera into a special wireless mode, wait for it to find a hotspot, go through the various "wizard" steps to authenticate, select your pics and then wait while they uploaded to the Kodak site. All while dodging the rubber bullets, water cannons and baton charges.
It would be faster and more effective to hand the card over to someone else who retreats from the front lines while you take another set of pics. Alternatively if you absolutely had to broadcast pics yourself, you'd be better off to do it from a pocket pc and software that dumps the entire contents of an inserted card to the net.
I wouldn't diss the idea completely - after all if your camera would connect to an ad-hoc network you could perhaps arrange for someone with a PDA or small laptop to shadow you at some distance and broadcast the pics back to them, but it would still be an awkward arrangement. And its doubtful that this camera would help you do that.
Perhaps it's simpler and equally effective to use redundancy - multiple photographers, with each passing their filled memory cards to runners.
That it only connects to Kodak's own shitty gallery service?
It makes more sense to do it this way than implement PDF into every app since a printer driver works with anything. For example I print receipts out from Firefox using the PDF printer.
For example, if it costs $5 million as speculated by some official to move to OpenOffice and $50 million to update / retrain for MS Office 12, how much would it cost to stay put with the existing MS Office and spend money developing ODF filters for it? It must be doable, in fact it is doable. Microsoft have implied as much when they said they would develop read / write filters for their XML format in versions of Office going back to 2000. If they can do their format, then ODF must be a shoe-in too.
I'd rather see that happen (assuming they released the code afterwards) than see them go to OpenOffice. A filter for MS Word & Excel would be immeasurably more valuable than a couple of thousand OO seats.
I know about this, but the exchange plugin needs to be part of it too. It doesn't make any difference for home use but a calendar / email app which doesn't talk to Exchange is next to useless in a lot of corporate environments. I am actually surprised that someone with deep pockets like IBM hasn't thrown a ton of cash as bounty for whoever integrates the exchange plugin into Thunderbird.
It makes a change from stories being lifted straight from Fark.
The only reason it even exists as far as I can make out is because once a company gets MS Exchange, it's about the only client that is designed to work with it.
Personally I'd love to see Evolution on Windows. Its presence might shake the tree a bit. Even better (for me) would be to see Sunbird / Thunderbird merged and using the Novell Exchange plugin. I think Thunderbird is a killer email app, but the lack of Exchange support hurts it in the business environment.
I not really familiar with the eDonkey network, was eDonkey packed with spyware? It would explain a lot if it was. I know Kazaa was too which prompted someone to produce a Kazaa Lite with all the spyware, ads, cookies, trackers, affiliate links and everything else ripped out.
I thought the open source and decentralized eMule was the tool of choice for the eDonkey network, with Shareaza and other tools following closely behind.
That's better than nothing, but really the API to the theme engine should be a simple set of C APIs that can be implemented very easily and loaded against GTK and QT. This API and its headers would be specified by freedesktop and authors just have to implement their themes against the API. When GTK / QT loads the engine, it enumerates all themes, loads the user's selected engine and calls the functions to draw buttons, get metrics, get system colours and everything else that the theme should do.
It shouldn't be necessary for GTK to piggyback over QT or vice versa - just define a nice spec which allows both widget sets to call a well defined API.