If it took 'em 6000 pages to describe the useless OOXML spec, I wouldn't count on these 14000 being all that useful either (with or without patent roadblocks).
That's a silly comparison. Nobody viewing normal websites keeps their pipe in constant use. Presumably you take some time to actually look at the sites you download. I don't think they'd be too happy if you ran an automated web crawler over your home circuit either.
ISP's are ultimately going to have to go to a model like cellphone contracts. 100 GB per month (or whatever). After that your bandwidth drops off or you pay for the overage, depending on your plan. Carry-over MB's and all. At least that's nominally fair. And maybe for those non-downloaders, there'd be a really low-cost, low volume plan.
As it stands, the new OOXML 'standard' amounts to a mandate to upgrade to Office 2007 (yes, there's some kind of add-on for older versions, but most will just eat the upgrade). A nice win for MS.
It would be nice if Government mandates required that multiple, compatible implementations exist for whatever standards they mandate.
That might call Microsoft's bluff. Either they'd have to implement a working OOXML to ODF translator or help others implement OOXML and verify completeness.
Hell, by defining 'standard' in terms of actual multiple implementations, Office 2000 binary would make a better standard than OOXML. OOo does a pretty good job of reading them - better than anybody but MS is likely to do for OOXML anytime soon.
So, let's lobby for governments to just standdardize on ODF, PDF and Office 2000.
Of course, Abiword, KOffice and OOo would have to get cracking on making their ODF implementations compatible for ODF to make the cut.
Ummm.... Because that competing poprietary technology has been adopted as an industry standard by the media companies whose content is to be delivered that way.
The only reason in hell why they'd consider sticking with their own competing proprietary tech is that, once again, they think they can still use their desktop monopoly to make it a success.
It seems like if Google can get WINE to the point that Photoshop can run well, Adobe could at least make a winelib build.
Isn't it possible to get more native behavior by building a wine-aware binary that knows enough to not use drive letters, etc.
Or is Adobe convinced nobody'd use it even if they did. Presumably Google's done some research to determine that there's demand for this - unless they just want to help improve WINE.
How quickly they forget. The Bush Administration argued that taxing those dividends amounted to double taxation too. So let's see. If you tax the profits, it's double taxation with the dividends. If you tax the dividends, it's double taxation with the profits. So let's tax neither. Total bullshit.
In the name of states' rights, we've set up a bunch of banana republics that operate largely as tax havens for one industry or another. It seems to me to be less a question of the morality of taxing corporations, but of the practicality of finding a foolproof way to do it. Maybe multistate corporations should pay their taxes to the feds and then have the proceeds divvied up among all the states. That way, business could locate wherever it makes sense, and we'd all reap the benefits of their share of the cost of running a government.
And if you think there shouldn't be a government to pay for, well this is a democracy. Try voting that in. Good luck.
Sure, within reason. But assuming bandwidth is a finite resource it may not be practical to allow everybody to download all the huge files they want for one low price. Maybe it is practical, and the issue is just ISP greed, but I'm assuming for the sake of argument that ISP greed isn't the whole story.
It seems to me that 20GB per month worth of downloading big files fast ought to be plenty for most of us. About a Linux ISO per day, I'd think. And that's my point. We download this stuff because it costs us exactly nothing. If it cost me a few bucks to download, I might go to Cheap Bytes or somewhere instead for a CD. And that wouldn't be the worst thing in the world. The internet is a disruptive technology alright, but it may be being made artificially so due to stuff like moratoriums on sales tax, loss-leader free shipping, etc. I'm not (entirely) sure that I don't want there to still be a neighborhood bookstore, and if it's killed because it can't compete with, say, Amazon - and that's because Amazon customers can afford the convenience because the government said there should be no sales tax (for now), that's not as great a deal as it may sound.
I pay for broadband mainly so that day to day web browsing (well within the suggested 20GB limit) is fast. Not so I can download music and movies, etc. It's nice that I can, but given the choice I might prefer to forgo that for the option of cutting my broadband bill. Of course, nobody's offering to cut my broadband bill, so the whole exercise is probably moot. Still, the point is that broadband costs something, and the pricing structure needs to reflect that. Unlimited downloads at a low, low price is not likely to fit into a reasonable pricing structure if you expect capacity to keep up with demand.
If bandwidth usage is really an economic issue for ISP's, then there are several ways to deal with that fairly. A bandwidth cap, as long as it's reasonable (20GB seems pretty reasonable to me) would be preferable to throttling iTunes, YouTube, or porn for that matter. Personally, I'd prefer pricing tiers based on traffic, not speed. You pay for some amount of traffic, and then pay more if you go over. Either way, what you access, you get as fast as possible.
The point is to let the customer decide what they want to access. If it costs a dollar or two to download the equivalent of a CD, maybe you should buy the CD and use your bandwidth for something else (or just save the money and pay less for Internet access). Maybe that'd get the RIAA off our backs. In any case, don't tell me what I can access.
I guess there's a flip side to that. If the content's something like on-demand video rental, why shouldn't your ISP be able to provide a cheaper service based on having direct connectivity to you? Or, put another way, if it costs them less to deliver the service to you, why shouldn't you reap some of the savings? There are different aspects to the net neutrality issue. A Netflix isn't providing original content. We may gain in terms of competition based on their having, essentially, a free delivery mechanism though. I guess that's good for us and good for Netflix, but it's obviously not good for our ISP's. Do we care about that? I'm not sure, but there are two sides to the issue.
I agree that being able to pause and rewind a program while you're still recording it is Tivo's big innovation. But essentially that's an obvious extension of their original innovation, recording to a disk instead of a tape. The random-access nature of a disk makes this a trivial 'achievement'. It's still clever as hell to think of it in the first place, but obvious to figure out how to do it once you're recording to a random-access device.
In a way, this is similar to Amazon's one-click patent. Combine a few readily available components to produce a clever new benefit, which just happens to be a natural extension of what those components always did.
So the real question is - are patents supposed to reward clever ideas, or are they supposed to make public mechanisms for which it would be hard to answer the question "how did they do that?" without their telling you? I guess Tivo might have had to tell you there's a hard disk inside the box, but once you know that, the rest is obvious. Still, this is a lot more innovative than one-click. Anyone could figure out that one-click is just another way to use a cookie.
No. ODF is really about lowering the cost of desktop productivity software.
That's certainly good for the public. Not necessarily so great for IBM, except that it levels the (hopefully, soon to be lower) playing field.
I've been waiting for *someone* to buy TrollTech..
on
Nokia Buys Trolltech
·
· Score: 4, Interesting
I'm personally crossing my fingers for Nokia to change the license to LGPL.
Nokia doesn't make their money licensing software, and I don't think they want to change that now. TrollTech was relatively cheap, because there wasn't a really lucrative market for their commercial licenses. TT had to stick with the dual-license model, because they had no other revenue stream. Nokia is a hardware manufacturer, and I'd think running their hardware on a mainstream software platform would be important to them. Going LGPL would go a long way toward accomplishing that.
Unless Nokia fears their competitors having equal access to the same software platform, a move to the LGPL would be all to the upside. And if they do fear that, then they could fork the Qtopia phone platform and keep that GPL. Or even drop the GPL version and go completely commercial on that. But these days, smart phones need a developer-friendly platform every bit as much as desktop systems do. QT would have some performance advantages over Google's Java-based phone platform. And Nokia, as the first mover and primary maintainer of the platform, ought to be able to leverage that into a huge lead.
That's if they make the switch to LGPL. And if they don't? They'll have a great phone platform, but less open to 3rd party developers. If they think, based on that, they can win a competitive battle for setting smart phone standards with Google and Microsoft, go for it. But I don't think they can. They're smart. They understand why Linux has all the buzz , BSD does not, and OS/2 is gone. GPL for apps, LGPL for libraries. It's scary to a commercial enterprise, but it really works - at least better than anything else (except, maybe, having a monopoly on desktop operating systems...).
...what is it about Eclipse that makes it so slow? I gave the Lotus Symphony thing a try and thought - nice beginning, but if you can't make it faster, this isn't going to fly. (yes, fast dual-core processor - lots of memory - is 1GB still 'lots'?)
Is it Java? Is it the size of the toolkit? Or, in the case of Symphony, is it the fact that under all that bloat, you have the bloat of OpenOffice? OpenOffice (2.3, at least) is much snappier, though. I can forgive OpenOffice its long load times, since it's not noticeably sluggish once it's started. But Symphony takes forever to start and is then sluggish once its (admittedly pretty) interface is up and running. And it's compounded by their ambitious sidebar thing, which flips as you change context moving around your document, but doesn't keep up with your movements. Ends up being a distraction instead of a powerful interface paradigm (actually, I think it might even be distracting even if it did keep up).
I thought the point of Eclipse was, unlike Swing, to implement the toolkit natively on each platform. If so, it sounds like a great idea. Am I just seeing an interim step toward a toolkit that will eventually work like that?
I've even tried using the Eclipse IDE as a programmer's editor to work on unix source code from a Windows desktop via Samba. Admittedly overkill, but it was free, my company was slow in agreeing to pay for a commercial editor, and I was getting tired of vim (vim, unlike vi, is really slow for some reason on my old AIX box). Eclipse was better than I expected for this purpose (one of my programmers still uses and likes it), but no better than vim over telnet for my tastes. I'm actually hopeful at the prospect of using kate once KDE apps on Windows are stable.
Anyway, I digress. I applaud IBM for its support of Linux for its desktop applications. I'm just afraid that relying on Eclipse to do it might be a mistake. If only IBM would buy Trolltech, switch QT to the LGPL and open up another, perhaps more viable, option.
A final thought. Java, Eclipse (and.NET, for that matter) might make sense as a way to deliver binary portable apps in a vertial market where apps are very complex and constantly changing. Binary portability would be a huge boon to developers in such cases, assuming the vendor cares about portability in the first place. But for traditional productivity apps, I think the QT portability model probably works better. Those kinds of apps are more self-contained, typically more mature, and (let's face it) are competing with native apps (on the major platforms, at least).
Somebody - that people would trust to not be sending around viruses (Sun?, Google?) - ought to write a tiny downloadable app that will change your default format in Office 2007 back to.doc. Seriously, this.docx default is causing a lot of people problems, and not just ODF fans. You'd be surprised how many people can't figure out how to change the default. And without MS0 2007 as a reference, I can't walk some of the more literal users that end up asking through finding it in the entirely new menu system I've never seen (click File, click Print,... where's File?).
A nice little web link on google.com ("Are your friends complaining about not being able to open your Word 2007 documents? Fix it here") would do the trick.
Ok, Ok. I know 'choice is good' and we don't want a winner in the desktop wars. But the dream of a single Linux ABI to target still lives on in some of us. Unfortunately, for those of us who find KDE to be a superior desktop to the other choices, it's never gonna be part of a single ABI.
There will always be commercial apps, and because they don't provide re-compilable source code, they're always gonna want to target a single ABI. And they will always go for a less restrictive license. So, assuming we want those commercial apps, and assuming they'll go with something they can use for free, there will always be GNOME. Not so bad, assuming there can one day be 1 GNOME target and that the various tookits will one day integrate seamlessly. The wildcard is the Mac, because most ISV's would want to target it, and QT would be a natural choice if it were free.
I don't begrudge QT their business model. They are a business, they've made some great contributions, and they want to eat. Well. And why shouldn't they. But their chosen model - free for GPL'd stuff and (relatively) expensive for non-GPL'd stuff - limits their market. It's not as though there aren't other models. Microsoft charges a bundle for Visual Studio, and (more or less) requires it to develop with their otherwise free (to use) ABI. Maybe QT could produce a comparable - but multi-platform - IDE, and make their money selling that. Then the commercial ISV's could use QT without paying for VS too. The problem is that the commercial ISV's have already invested in VS and a lot of Windows-specific code. Unless they really want to target Macs or Linux, it's really hard for them to justify the expense of going portable with QT.
Or maybe some company (IBM - yeah, but why; Apple - yeah, but mustn't risk losing MSOffice; Sun - see IBM) could give the TrollTech'rs a big payday and set QT free (as in Beer). The trick is to find someone who could justify the price by selling lots of hardware. But the server vendors don't care so much about GUI stuff, and the desktop vendors don't sell hardware, except Apple (see above). Google'd have been a good choice if they'd gone with QTopia for their phone platform (though they won't be actually building and selling phone hardware, will they?).
I use Firefox on my PCLOS box to access Citibank online all the time. For some reason they don't like however my Firefox is identifying itself (pretty dumb, since they accept Firefox on Windows without a glitch). But at least, Citibank provides a 'proceed anyway' link that lets you in and everything works perfectly.
So, it's odd that this CitiCards site wouldn't work - unless they're blocking you at the doorway. It work with Firefox/Windows.
Proprietary software lends itself to the dubious practice of buying the competition in order to shut them down. If you can't compete, and you have enough money, you can in many cases legally put the competition out of business.
In theory, no company would do this to the better of two products, but in practice it works out differently. The company I work for has been on a buying binge for the last 10 years. They now own almost all of the products in several vertical markets. That is, all except the clear market leader in each. So what they really have is multiple sets of competing product lines, none of which is full-featured enough to go head-to-head with the market leader.
The next step is not to pick one and flesh it out. It's to buy yet another company with a vaporware product that sounds good on paper and flesh that out. They use all the revenue streams of the existing products to finance the new one. The problem is that the new one will never be a market leader (for various reasons related to the reason it was sold as vaporware in the first place).
The end result is that a bunch of products that were each at one time potential contenders have been crippled, stymied, and ultimatly, most of them will be shut down. And some of those products (mine, specifically) contain some really nice innovations. Innovations that would make them competitive with the new vaporware product - but the company needn't worry about that, because they get to decide which potential customers these products will be marketed to. And any customer that could possibly be served by the new product line will never see those innovations, so the marketplace will never get a chance to decide on their value.
A sad state of affairs, made worse by some weird quirk in antitrust law that allows you to buy a huge market share in small increments, where it would not allow you to merge the leaders into a similar near-monopoly. All I know is that there have been about 6 corporate owners over the lifetime of the product line I work on. What I don't understand is how, assuming much of US business works this way, anybody makes any money.
Remember when Judge Jackson told a reporter what he thought of Microsoft coming into his court with blatant lies and fake demonstrations. His ruling was overturned because of his *bias*.
It seems that if you present a laughably fraudulent case, and those evaluating that case can't hold their tongues about how laughably fraudulent it is, you win. Outrageous dishonesty pays off better than simple dishonesty.
I was hoping somebody would bring up multicast. I've seen the term bandied about in the past, and I assume it refers to an IP-based equivalent to TV broadcasting. Multiple people receive the same stream, giving up control over when they tune it. Certainly makes sense for 'interntet TV' - especially if you could TIVO the stream to get time-shifting capability that way.
Has this actually been designed, or is it just something people talk about? Anybody have a URL that goes into detail?
I think what the guy's talking about is the fact that the system was designed with a certain ratio of capacity between the backbone and the end-users' connections. If everybody has megabits of bandwidth, and they all want to use all of it all the time, you need a huge capacity backbone.
It doesn't matter whether you're talking about 16k baud rates or 10 Megabit connections, if all that data has to go through a bottleneck that's not many times the capacity of the individual connections, you're gonna run out.
In a normal market, they shouldn't. As long as they can charge you for XP, why should Microsoft care whether you bought XP from them or Vista? In fact, if you buy XP now, MS ought to be able to get more from you for a future upgrade if and when they come out with something that's actually better.
But XP's not in sync with their agenda. There's stuff in Vista that's there in order to bolster Microsoft's position in web search. And stuff to bolster.NET as an application platform. In other words, they want to sell you Vista instead of XP, because that's how they plan to bundle their way into future dominance in other arenas. Of course, they can provide the same 'features' as upgrades to XP, but then they'd have to get people to install the upgrades. That's (part of) the difference between monopoly bundling and normal business. The monopoly folks can go to third parties and say "write.NET apps, because 90+% of PC's will be able to run them off the shelf". As opposed to "write.NET apps, because they work great". The 'path of least resistance' argument is powerful, indeed.
Problem is, they blew it. The hardware requirements of Vista have more or less guaranteed that there's no upgrade stream from XP to Vista on older boxes. That's lost revenue and lost 'inevitability' points. But as long as XP can be made to conform to their 'vision', this shouldn't be too much of a problem for them.
...how come all the iPod wannabes support WMA but not non-DRM AAC. Most iPod owners' collections consist primarily of AAC's ripped from CDs. Why on earth would the 2nd tier players not want to be able to play these, if only to lower that barrier to entry?
Do they all think that hitching their wagons to Microsoft (and MS DRM) will magically win the day for them? Even now? I know Apple won't let them use the iPod's DRM, which I guess is pretty nasty. But that's no reason to snub Apple customers willing to switch players if not for having to re-rip their collections.
He's probably right that the GPL3 closes a dangerous loophole, but that's not what Linus objects to. It's the anti-Tivoization clause Linux objects to, and that has nothing to do with this loophole... unless you think that the computer industry is going to produce Linux desktop computers that cannot be run without a Tivo-like lock in the kernel - and that people will actually buy those computers.
Well, maybe that's not as farfetched as it sounds - but more likely, those locked down computers would be Windows-only boxes. Even so, why not make a narrower clause that prevents locking down the kernel in a *general-purpose* computer. What Tivo does is build a special-purpose *appliance* that happens to be a computer. The fact that they run it on Linux is a good thing. The fact that they don't let you buy one of their subsidized boxes and not pay for the service that subsidizes it may not be ideal, but does nothing to inhibit the success of the Linux software they use. That's where Linux gets it better than RMS. He's willing to give up control of how his code is used, and appreciates that if he didn't do that, the software would have withered on the vine. If Tivo is a freeloader (and I'm not saying they are), Linus doesn't care - it doesn't hurt him or Linux.
It's Linux vs Hurd, and it's pretty obvious that Linux wins - unless you're typing your posts from a Hurd box.
Slashdot Mirror
If it took 'em 6000 pages to describe the useless OOXML spec, I wouldn't count on these 14000 being all that useful either (with or without patent roadblocks).
That's a silly comparison. Nobody viewing normal websites keeps their pipe in constant use. Presumably you take some time to actually look at the sites you download. I don't think they'd be too happy if you ran an automated web crawler over your home circuit either.
ISP's are ultimately going to have to go to a model like cellphone contracts. 100 GB per month (or whatever). After that your bandwidth drops off or you pay for the overage, depending on your plan. Carry-over MB's and all. At least that's nominally fair. And maybe for those non-downloaders, there'd be a really low-cost, low volume plan.
As it stands, the new OOXML 'standard' amounts to a mandate to upgrade to Office 2007 (yes, there's some kind of add-on for older versions, but most will just eat the upgrade). A nice win for MS.
It would be nice if Government mandates required that multiple, compatible implementations exist for whatever standards they mandate.
That might call Microsoft's bluff. Either they'd have to implement a working OOXML to ODF translator or help others implement OOXML and verify completeness.
Hell, by defining 'standard' in terms of actual multiple implementations, Office 2000 binary would make a better standard than OOXML. OOo does a pretty good job of reading them - better than anybody but MS is likely to do for OOXML anytime soon.
So, let's lobby for governments to just standdardize on ODF, PDF and Office 2000.
Of course, Abiword, KOffice and OOo would have to get cracking on making their ODF implementations compatible for ODF to make the cut.
Any guesses which job would be easier?
Ummm.... Because that competing poprietary technology has been adopted as an industry standard by the media companies whose content is to be delivered that way.
The only reason in hell why they'd consider sticking with their own competing proprietary tech is that, once again, they think they can still use their desktop monopoly to make it a success.
It seems like if Google can get WINE to the point that Photoshop can run well, Adobe could at least make a winelib build.
Isn't it possible to get more native behavior by building a wine-aware binary that knows enough to not use drive letters, etc.
Or is Adobe convinced nobody'd use it even if they did. Presumably Google's done some research to determine that there's demand for this - unless they just want to help improve WINE.
Half of the shit in windows/zune/xbox/server/sync..... was in some way created or innovated or copied at MSR.
Seriously, what's innovative about zune/xbox/server/sync... or windows, for that matter?
How quickly they forget. The Bush Administration argued that taxing those dividends amounted to double taxation too. So let's see. If you tax the profits, it's double taxation with the dividends. If you tax the dividends, it's double taxation with the profits. So let's tax neither. Total bullshit.
In the name of states' rights, we've set up a bunch of banana republics that operate largely as tax havens for one industry or another. It seems to me to be less a question of the morality of taxing corporations, but of the practicality of finding a foolproof way to do it. Maybe multistate corporations should pay their taxes to the feds and then have the proceeds divvied up among all the states. That way, business could locate wherever it makes sense, and we'd all reap the benefits of their share of the cost of running a government.
And if you think there shouldn't be a government to pay for, well this is a democracy. Try voting that in. Good luck.
Sure, within reason. But assuming bandwidth is a finite resource it may not be practical to allow everybody to download all the huge files they want for one low price. Maybe it is practical, and the issue is just ISP greed, but I'm assuming for the sake of argument that ISP greed isn't the whole story.
It seems to me that 20GB per month worth of downloading big files fast ought to be plenty for most of us. About a Linux ISO per day, I'd think. And that's my point. We download this stuff because it costs us exactly nothing. If it cost me a few bucks to download, I might go to Cheap Bytes or somewhere instead for a CD. And that wouldn't be the worst thing in the world. The internet is a disruptive technology alright, but it may be being made artificially so due to stuff like moratoriums on sales tax, loss-leader free shipping, etc. I'm not (entirely) sure that I don't want there to still be a neighborhood bookstore, and if it's killed because it can't compete with, say, Amazon - and that's because Amazon customers can afford the convenience because the government said there should be no sales tax (for now), that's not as great a deal as it may sound.
I pay for broadband mainly so that day to day web browsing (well within the suggested 20GB limit) is fast. Not so I can download music and movies, etc. It's nice that I can, but given the choice I might prefer to forgo that for the option of cutting my broadband bill. Of course, nobody's offering to cut my broadband bill, so the whole exercise is probably moot. Still, the point is that broadband costs something, and the pricing structure needs to reflect that. Unlimited downloads at a low, low price is not likely to fit into a reasonable pricing structure if you expect capacity to keep up with demand.
If bandwidth usage is really an economic issue for ISP's, then there are several ways to deal with that fairly. A bandwidth cap, as long as it's reasonable (20GB seems pretty reasonable to me) would be preferable to throttling iTunes, YouTube, or porn for that matter. Personally, I'd prefer pricing tiers based on traffic, not speed. You pay for some amount of traffic, and then pay more if you go over. Either way, what you access, you get as fast as possible.
The point is to let the customer decide what they want to access. If it costs a dollar or two to download the equivalent of a CD, maybe you should buy the CD and use your bandwidth for something else (or just save the money and pay less for Internet access). Maybe that'd get the RIAA off our backs. In any case, don't tell me what I can access.
I guess there's a flip side to that. If the content's something like on-demand video rental, why shouldn't your ISP be able to provide a cheaper service based on having direct connectivity to you? Or, put another way, if it costs them less to deliver the service to you, why shouldn't you reap some of the savings? There are different aspects to the net neutrality issue. A Netflix isn't providing original content. We may gain in terms of competition based on their having, essentially, a free delivery mechanism though. I guess that's good for us and good for Netflix, but it's obviously not good for our ISP's. Do we care about that? I'm not sure, but there are two sides to the issue.
I agree that being able to pause and rewind a program while you're still recording it is Tivo's big innovation. But essentially that's an obvious extension of their original innovation, recording to a disk instead of a tape. The random-access nature of a disk makes this a trivial 'achievement'. It's still clever as hell to think of it in the first place, but obvious to figure out how to do it once you're recording to a random-access device.
In a way, this is similar to Amazon's one-click patent. Combine a few readily available components to produce a clever new benefit, which just happens to be a natural extension of what those components always did.
So the real question is - are patents supposed to reward clever ideas, or are they supposed to make public mechanisms for which it would be hard to answer the question "how did they do that?" without their telling you? I guess Tivo might have had to tell you there's a hard disk inside the box, but once you know that, the rest is obvious. Still, this is a lot more innovative than one-click. Anyone could figure out that one-click is just another way to use a cookie.
No. ODF is really about lowering the cost of desktop productivity software.
That's certainly good for the public. Not necessarily so great for IBM, except that it levels the (hopefully, soon to be lower) playing field.
I'm personally crossing my fingers for Nokia to change the license to LGPL.
Nokia doesn't make their money licensing software, and I don't think they want to change that now. TrollTech was relatively cheap, because there wasn't a really lucrative market for their commercial licenses. TT had to stick with the dual-license model, because they had no other revenue stream. Nokia is a hardware manufacturer, and I'd think running their hardware on a mainstream software platform would be important to them. Going LGPL would go a long way toward accomplishing that.
Unless Nokia fears their competitors having equal access to the same software platform, a move to the LGPL would be all to the upside. And if they do fear that, then they could fork the Qtopia phone platform and keep that GPL. Or even drop the GPL version and go completely commercial on that. But these days, smart phones need a developer-friendly platform every bit as much as desktop systems do. QT would have some performance advantages over Google's Java-based phone platform. And Nokia, as the first mover and primary maintainer of the platform, ought to be able to leverage that into a huge lead.
That's if they make the switch to LGPL. And if they don't? They'll have a great phone platform, but less open to 3rd party developers. If they think, based on that, they can win a competitive battle for setting smart phone standards with Google and Microsoft, go for it. But I don't think they can. They're smart. They understand why Linux has all the buzz , BSD does not, and OS/2 is gone. GPL for apps, LGPL for libraries. It's scary to a commercial enterprise, but it really works - at least better than anything else (except, maybe, having a monopoly on desktop operating systems...).
...what is it about Eclipse that makes it so slow? I gave the Lotus Symphony thing a try and thought - nice beginning, but if you can't make it faster, this isn't going to fly. (yes, fast dual-core processor - lots of memory - is 1GB still 'lots'?)
.NET, for that matter) might make sense as a way to deliver binary portable apps in a vertial market where apps are very complex and constantly changing. Binary portability would be a huge boon to developers in such cases, assuming the vendor cares about portability in the first place. But for traditional productivity apps, I think the QT portability model probably works better. Those kinds of apps are more self-contained, typically more mature, and (let's face it) are competing with native apps (on the major platforms, at least).
Is it Java? Is it the size of the toolkit? Or, in the case of Symphony, is it the fact that under all that bloat, you have the bloat of OpenOffice? OpenOffice (2.3, at least) is much snappier, though. I can forgive OpenOffice its long load times, since it's not noticeably sluggish once it's started. But Symphony takes forever to start and is then sluggish once its (admittedly pretty) interface is up and running. And it's compounded by their ambitious sidebar thing, which flips as you change context moving around your document, but doesn't keep up with your movements. Ends up being a distraction instead of a powerful interface paradigm (actually, I think it might even be distracting even if it did keep up).
I thought the point of Eclipse was, unlike Swing, to implement the toolkit natively on each platform. If so, it sounds like a great idea. Am I just seeing an interim step toward a toolkit that will eventually work like that?
I've even tried using the Eclipse IDE as a programmer's editor to work on unix source code from a Windows desktop via Samba. Admittedly overkill, but it was free, my company was slow in agreeing to pay for a commercial editor, and I was getting tired of vim (vim, unlike vi, is really slow for some reason on my old AIX box). Eclipse was better than I expected for this purpose (one of my programmers still uses and likes it), but no better than vim over telnet for my tastes. I'm actually hopeful at the prospect of using kate once KDE apps on Windows are stable.
Anyway, I digress. I applaud IBM for its support of Linux for its desktop applications. I'm just afraid that relying on Eclipse to do it might be a mistake. If only IBM would buy Trolltech, switch QT to the LGPL and open up another, perhaps more viable, option.
A final thought. Java, Eclipse (and
Somebody - that people would trust to not be sending around viruses (Sun?, Google?) - ought to write a tiny downloadable app that will change your default format in Office 2007 back to .doc. Seriously, this .docx default is causing a lot of people problems, and not just ODF fans. You'd be surprised how many people can't figure out how to change the default. And without MS0 2007 as a reference, I can't walk some of the more literal users that end up asking through finding it in the entirely new menu system I've never seen (click File, click Print,... where's File?).
A nice little web link on google.com ("Are your friends complaining about not being able to open your Word 2007 documents? Fix it here") would do the trick.
...unless they're taking huge tax deductions on what is essentially money spent to buy marketshare.
Ok, Ok. I know 'choice is good' and we don't want a winner in the desktop wars. But the dream of a single Linux ABI to target still lives on in some of us. Unfortunately, for those of us who find KDE to be a superior desktop to the other choices, it's never gonna be part of a single ABI.
There will always be commercial apps, and because they don't provide re-compilable source code, they're always gonna want to target a single ABI. And they will always go for a less restrictive license. So, assuming we want those commercial apps, and assuming they'll go with something they can use for free, there will always be GNOME. Not so bad, assuming there can one day be 1 GNOME target and that the various tookits will one day integrate seamlessly. The wildcard is the Mac, because most ISV's would want to target it, and QT would be a natural choice if it were free.
I don't begrudge QT their business model. They are a business, they've made some great contributions, and they want to eat. Well. And why shouldn't they. But their chosen model - free for GPL'd stuff and (relatively) expensive for non-GPL'd stuff - limits their market. It's not as though there aren't other models. Microsoft charges a bundle for Visual Studio, and (more or less) requires it to develop with their otherwise free (to use) ABI. Maybe QT could produce a comparable - but multi-platform - IDE, and make their money selling that. Then the commercial ISV's could use QT without paying for VS too. The problem is that the commercial ISV's have already invested in VS and a lot of Windows-specific code. Unless they really want to target Macs or Linux, it's really hard for them to justify the expense of going portable with QT.
Or maybe some company (IBM - yeah, but why; Apple - yeah, but mustn't risk losing MSOffice; Sun - see IBM) could give the TrollTech'rs a big payday and set QT free (as in Beer). The trick is to find someone who could justify the price by selling lots of hardware. But the server vendors don't care so much about GUI stuff, and the desktop vendors don't sell hardware, except Apple (see above). Google'd have been a good choice if they'd gone with QTopia for their phone platform (though they won't be actually building and selling phone hardware, will they?).
Oh well.
And while we're at it, note that the Zunes recieved customer ratings of 3 1/2 stars vs 4 1/2 for the iPods. Sounds a bit fishy.
I use Firefox on my PCLOS box to access Citibank online all the time. For some reason they don't like however my Firefox is identifying itself (pretty dumb, since they accept Firefox on Windows without a glitch). But at least, Citibank provides a 'proceed anyway' link that lets you in and everything works perfectly.
So, it's odd that this CitiCards site wouldn't work - unless they're blocking you at the doorway. It work with Firefox/Windows.
Proprietary software lends itself to the dubious practice of buying the competition in order to shut them down. If you can't compete, and you have enough money, you can in many cases legally put the competition out of business.
In theory, no company would do this to the better of two products, but in practice it works out differently. The company I work for has been on a buying binge for the last 10 years. They now own almost all of the products in several vertical markets. That is, all except the clear market leader in each. So what they really have is multiple sets of competing product lines, none of which is full-featured enough to go head-to-head with the market leader.
The next step is not to pick one and flesh it out. It's to buy yet another company with a vaporware product that sounds good on paper and flesh that out. They use all the revenue streams of the existing products to finance the new one. The problem is that the new one will never be a market leader (for various reasons related to the reason it was sold as vaporware in the first place).
The end result is that a bunch of products that were each at one time potential contenders have been crippled, stymied, and ultimatly, most of them will be shut down. And some of those products (mine, specifically) contain some really nice innovations. Innovations that would make them competitive with the new vaporware product - but the company needn't worry about that, because they get to decide which potential customers these products will be marketed to. And any customer that could possibly be served by the new product line will never see those innovations, so the marketplace will never get a chance to decide on their value.
A sad state of affairs, made worse by some weird quirk in antitrust law that allows you to buy a huge market share in small increments, where it would not allow you to merge the leaders into a similar near-monopoly. All I know is that there have been about 6 corporate owners over the lifetime of the product line I work on. What I don't understand is how, assuming much of US business works this way, anybody makes any money.
Remember when Judge Jackson told a reporter what he thought of Microsoft coming into his court with blatant lies and fake demonstrations. His ruling was overturned because of his *bias*.
It seems that if you present a laughably fraudulent case, and those evaluating that case can't hold their tongues about how laughably fraudulent it is, you win. Outrageous dishonesty pays off better than simple dishonesty.
I was hoping somebody would bring up multicast. I've seen the term bandied about in the past, and I assume it refers to an IP-based equivalent to TV broadcasting. Multiple people receive the same stream, giving up control over when they tune it. Certainly makes sense for 'interntet TV' - especially if you could TIVO the stream to get time-shifting capability that way.
Has this actually been designed, or is it just something people talk about? Anybody have a URL that goes into detail?
I think what the guy's talking about is the fact that the system was designed with a certain ratio of capacity between the backbone and the end-users' connections. If everybody has megabits of bandwidth, and they all want to use all of it all the time, you need a huge capacity backbone.
It doesn't matter whether you're talking about 16k baud rates or 10 Megabit connections, if all that data has to go through a bottleneck that's not many times the capacity of the individual connections, you're gonna run out.
In a normal market, they shouldn't. As long as they can charge you for XP, why should Microsoft care whether you bought XP from them or Vista? In fact, if you buy XP now, MS ought to be able to get more from you for a future upgrade if and when they come out with something that's actually better.
.NET as an application platform. In other words, they want to sell you Vista instead of XP, because that's how they plan to bundle their way into future dominance in other arenas. Of course, they can provide the same 'features' as upgrades to XP, but then they'd have to get people to install the upgrades. That's (part of) the difference between monopoly bundling and normal business. The monopoly folks can go to third parties and say "write .NET apps, because 90+% of PC's will be able to run them off the shelf". As opposed to "write .NET apps, because they work great". The 'path of least resistance' argument is powerful, indeed.
But XP's not in sync with their agenda. There's stuff in Vista that's there in order to bolster Microsoft's position in web search. And stuff to bolster
Problem is, they blew it. The hardware requirements of Vista have more or less guaranteed that there's no upgrade stream from XP to Vista on older boxes. That's lost revenue and lost 'inevitability' points. But as long as XP can be made to conform to their 'vision', this shouldn't be too much of a problem for them.
...how come all the iPod wannabes support WMA but not non-DRM AAC. Most iPod owners' collections consist primarily of AAC's ripped from CDs. Why on earth would the 2nd tier players not want to be able to play these, if only to lower that barrier to entry?
Do they all think that hitching their wagons to Microsoft (and MS DRM) will magically win the day for them? Even now? I know Apple won't let them use the iPod's DRM, which I guess is pretty nasty. But that's no reason to snub Apple customers willing to switch players if not for having to re-rip their collections.
He's probably right that the GPL3 closes a dangerous loophole, but that's not what Linus objects to. It's the anti-Tivoization clause Linux objects to, and that has nothing to do with this loophole... unless you think that the computer industry is going to produce Linux desktop computers that cannot be run without a Tivo-like lock in the kernel - and that people will actually buy those computers.
Well, maybe that's not as farfetched as it sounds - but more likely, those locked down computers would be Windows-only boxes. Even so, why not make a narrower clause that prevents locking down the kernel in a *general-purpose* computer. What Tivo does is build a special-purpose *appliance* that happens to be a computer. The fact that they run it on Linux is a good thing. The fact that they don't let you buy one of their subsidized boxes and not pay for the service that subsidizes it may not be ideal, but does nothing to inhibit the success of the Linux software they use. That's where Linux gets it better than RMS. He's willing to give up control of how his code is used, and appreciates that if he didn't do that, the software would have withered on the vine. If Tivo is a freeloader (and I'm not saying they are), Linus doesn't care - it doesn't hurt him or Linux.
It's Linux vs Hurd, and it's pretty obvious that Linux wins - unless you're typing your posts from a Hurd box.