You may well be right (like I said, I'm not too familiar with MacOS), but since earlier versions of IE didn't support PNG at all, it is natural that people think it's IE that's to thank.
But whatever the solution is, interaction (with plugin/viewer) worked flawlessly; stacking transparent PNGs worked as one would expect, edge-smoothing using alpha too...
In any case it is kind of cool if IE is doing that. Not reinventing the wheel I mean.
Makes me wonder why Netscape can't use Quicktime on Mac as well?
ONe thing I forgot to mention, though, is that Mozilla (thank god) renders PNGs pretty decently as well, both on Mac, Windows and Linux. It would be a shame if Open Source application didn't support 'Open Format'.
That'd be a good justification for installing workstations in the bathrooms. Or maybe toilets at
the cubicles.
Hehe. Nice idea... But how well would it go with the "paperless office" ideas? About the first idea... umh... having work stations with Internet access in restrooms sounds kind of kinky.
I kind of agree that "ogg" doesn't really sound like an optimal "cool" name... But I don't see mp3 being much better ("what was that format... c3po?"); it's just that it has become common knowledge by becoming popular enough. Same for "gif" ("png" has much more kick in it doesn't it?)
It would be a definite plus to have a catchy name but weird names haven't really stopped other formats from becoming de facto standards. I don't think Linux, BSD, Gimp, gcc,.wav,.jp(e)g or dozens of other names/abbreviations had high coolness factor (ok some of those are still 'geek-only')
Hmmh. Wonder how much better would Linux have done if it had been named "Penguin"? ("Just Pengo for friends!") 8-)
This is an important point... Although I might say "people want to share but may also be willing to pay". In any case, I think that if ogg gains more popularity, apps will also become more sophisticated. Kind of inverse catch-22 situation.
So... I for one have been encoding in ogg for past 3 months (and will continue doing so), and if enough people do the same, ogg will become popular enough, more apps will support it, and level of top ogg-aware applications will rise to the level of top mp3-aware applications (most likely as those will then be the same apps...).
It would be interesting to know, though, how will ogg do in 'non-mp3' areas. I understood that it was at some point sponsored by internet-broadcasting company, ie. it was (also) meant to be a good streaming audio format. Becoming popular (alternative for RealAudio?) there shouldn't hurt its popularity in more 'static' use either.
IE 5 for Mac most certainly does show PNGs. And what is puzzling is that it has by far the best PNG support (better than IE on Windows).
I'm not a Mac-user, but I did develop PNG export functionality to a large shrink-wrapped application, and had to test it on various platforms. IE 5/Mac was surprisingly good, even doing correct gamma correction.
Also, if I recall correctly, NS 4.7x did show PNGs on Mac, albeit as badly as on any other platform (I don't have a mac to check this on right now).:-/
It is a pity in any case that NS fails miserably with transparency (except for 8-bit simple transparency images). IE on Windows is slightly better, but can't do 24-bit full alpha either.
Like others have pointed out (and especially when that 'review' was posted), test was either done without a clue, or the article omitted important properties. Anyways, here's the obligatory karma-whoring list of important things that weren't specified:
Bit rates used. In and of itself may render test completely meaningless.
Was it a true blind test or did reviewers know beforehand which format was which?
Which compression tools were used? For ogg this doesn't really matter, for mp3 it definitely does.
You have valid points about why it's not all that likely Palm would be buyer. However, I think moderators thought it was genuinely interesting, even if not realistic or very insightful? And no, obviously I wasn't moderating it.:-)
100-to-1 against Sun. Although they hate the Windows monopoly, they've raised their hands there... And concentrate on server side, enterprise computing etc. They have the money, but there's probably no place for BeOS in their strategy (unfortunately...). Would have been interesting to know what-could-be - case if they had bought Commodore (meaning Amiga) way back when
(which they were considering).
Well, I'm not a huge Qwest fan, but I had absolutely no problem signing up for Qwest DSL
but using another ISP (I know, YMMV).
Perhaps it is/was easier when you don't have to switch from Qwest.net, but it seemed to me that Qwest isn't all that hostile. Makes sense to me, because their DSL profits are most likely higher than their ISP profits (if they do any profit... it's much more customer service than physical DSL, and much smaller margins).
Anyways; I'd say that even with trouble, go with an external ISP (instead of Qwest.net). It is worth it.
I have Qwest's DSL (pro? 30$/month, 512 kbps) but a smaller ISP.
I've been pleased with both, and (I hope) am not in danger of being transferred to the Borg.:-)
People should perhaps consider using separate company for their physical line (DSL) and actual Internet services (ISP). Combining those might produce small savings (for me about 5$... but no static IP, just 2 email addresses, no servers allowed etc. etc.), but it might not be worth the hassle. Since Qwest is cool enough (or non-monopolistic... as unbelievable as it is for ex-US West) to allow 'competing' ISPs, why not use this option? Competition really is the key. Besides, personally I think Qwest should be making much much more profit on providing the low-level connection and letting outside ISP take care of costy end-user help desk. I mean; ISP is the place people first contact; and they only get (in my case) 40% of the cake...
Disclaimer: I don't know if this is possible everywhere Qwest has DSL-service. I know it is in Colorado, though.
The difference is that real operating systems (i.e., *nix), prevent ordinary user accounts from getting to the really low level / powerful things, like SUID programs or raw sockets.
However, installing linux on your PC is easy enough that you can get your raw socket access easy enough from home. And people who hack other machines to do spoofing usually get to root if they get any normal user account.
It is futile trying to prevent these kinds of access from single user desk top systems.
My experience has been that it's either a breeze (no probs whatsoever), or gets reasonably tricky. I've only installed linux on 6 different systems configs in past few years (ie. no statistical significance), but I'd say there's 50-50 chance. Of course, I bought all systems from no-name small companies, and in only 1 case the company chose the components 'guaranteed' to be linux-compatible (ie. they knew it'd be a linux box), so if big name vendors actually do/did check compatibility for their setups, it's more likely to be the "breeze" install.
But like others have said, the biggest problem really is that many people find installing _any OS_ difficult. And most of them need NOT install anything; PCs usually come with some flavour of Windows installed by default. Default installation of Red Hat (and I'm sure others like Mandrake, S.U.S.E) are really trivially easy (as long as h/w really is 'old enough' to be fully supported), and comparable to Windows ME install (which I had to re-install, thanks to the seller having partitioned 30 gig HD in one partition... ugh).
After installing Red Hat (for example), updating system is as easy as with windows, if you use RH's update agent (compared to windowsupdate.com).
--
Part of the alleged problems is that the other company is not obeying the restrictions of (Open/)Free Software license in question (GPL).
Since companies are involved, there are more relevant issues (relevant to the law suit(s)) contractual issues but I'm sure you can read about those in the article linked to as well as from previous./ - articles about the case.
--
Well, I'm not going to go through the whole "free" vs. "Free" lecture. Suffice to say that "freedom" doesn't have to refer to monetary payments.
But more importantly, if you sell your house for $200.000, and only get $45.000 will you be happy that you got "lots of money" for that? Or will you perhaps be mad at getting screwed? If the deal was for 2.5 M U$ then 300 k U$ really is "only"?
--
So, instead of trying to give information to its perhaps most important (potential) client group, they should just let misinformation and FUD fly? I actually fear you might be right, though. And I can easily imagine that lawyers would (or have) advice against trying to speak up, provide any truths that were not directly usable in lawsuit and so forth.:-/
It's just that it depends on what do you want more; win the legal cases, or try to stay in terms with developers. Perhaps you can't do both... Just sounds like if I had to do that, I'd feel dirty as hell, bending over and letting lawyers run the show, just like Evil Big Companies do. Even if I knew it may be the smartest thing to do.
--
If this is true, MySQL itself may be breaking other GPL licenses, or may not have a valid
license.
Well, if MySQL AB employs the actual authors/copyright holders of MySQL (like I think it does?), they can dual-license their code (triple license etc) however they want? Hell, they could create "users have to wear rainbow coloured donkey hats every friday if they use our product" license for NuSphere.:-)
If they were/are using GPL'ed code created by other people/companies, this would of course be different... But I think that's more what NuSphere is doing, and having problems with.
You don't see license infringement an issue? As in, "it's perfectly ok to disregard licensing"? Or as in "you haven't proved it happened"? Also, even if it's not part of the lawsuit, in OS/FS world it is relevant for developers, especially in trying to understand where the situation comes from. If the accusations (related to lawsuit specifics) seem petty, it may be that the history of the companies provides some more insight into why the lawsuits were filed.
In one of the previous threads related to the ongoing battle, there was an interesting note by someone familiar with history of NuSphere (ex-Progress). Basically what the writer said was that NuSphere most likely had/has problems dealing with Open/Free S/W since it used to be more traditional properietary code company. That probably explains many problems and (potential) misunderstandings.
I still can't blame MySQL AB for trying to make sure game rules are obeyed. It may have come as a surprise to NuSphere what exactly they get and what they don't get (ie. they weren't given a blanket "discard GPL restrictions for your product" rights... those could be given by dual licensing), but like they say, "ignorancy is no excuse"?
--
As I read the comment, if GPL is found to be n/a or invalid, it doesn't mean _no restrictions_ apply to the code. In fact, the code's original author(s) would then probably have full copyright control over the code. So, everything GPL _grants_ would be nullified...
And this, like he said, is probably even worse for than having to live with full restrictions GPL would impose.
--
Well, I probably should have clarified that I mean security patch - type upgrades. And yes, on work station - side, things get pretty obsolete. But for servers, although (large) companies dislike upgrading to new versions, they do usually apply the patches as required. Or perhaps that's only for the ones that are more mission critical than others, my view may be bit distorted. I'm sure mom and pop - ISPs are different from Fortune 500 dinosaurs.
--
Ah, those lil bastards! Wonder where I could order more of them?:-)
(yes, you got me, my native language has no articles... tricky bastards I'll tell you!)
--
...
Think about this for a minute. It is easy to conceive of ways in which much more damage could
be done to the internet than has already been done. If I recall correctly, the ILOVEYOU virus
deleted jpgs from hard drives. The worst results I am aware of from this is a commerical image
database being wiped out.
Commercial companies make backup copies of their databases on regular basis. This means that while financially losses are big (big bucks DB admins get paged to restore stuff from backups), the content losses are likely to be at most as severe as results of hardware failures (which do happen... even with sophisticated raid systems etc).
Also, unlike ignorant end users, companies (their admins) usually try to keep their systems up-to-date, and are likely to be less vulnerable to attacks. Even though as targets they are more visible, and probably more lucrative, too, they are much better prepared against threats than your regular Joe Sixpack.
Up until now, we have delt mainly with simple scripts whose workings are obvious.
I don't want to flame you here (you did say you are not a security expert), but usually worms are not just simple scripts (nor even non-word viruses); on unix-systems they may (and have) been scripts to be more portable, but there isn't anything simple in them either.
As to email being required... for decades (since first worms were created, early 80s?) worms have been able to use other network connections than email. That's the case with CR; variety is good for viruses and worms. Spreading using attachments is easy (some might say lame...) way to spread, but bit too obvious. Easy to implement, though, which is why it has been a popular approach.
I guess I just disagree with doomsday prophecies like this. Even though I don't want to appear like a MS-bashing zealot, I must say that Microsoft is now paying for putting security related issues on rather low priority for years. There's a lot that have been done by other companies and organizations (Java-security model by Sun, xBSD code inspections to build reasonably secure server OSes, etc. etc); Microsoft just didn't think potential risks were big enough. They have been proven wrong... and hopefully have started paying more attention.
--
BIND has got to be the closest open source product to IIS with respect to massive numbers
of vulnerabilities that give "immediate root access" to quote SANS.
Wonder if sendmail has lost its position as number one Intruder Service, then?:-)
--
While it is true that one shouldn't just assume VoIP uses 'stock' internet backbone connections, it may well be more cost-effective for carriers to mix traffic at some level. And if the mixing is done dynamically (ie. no hard-coded fixed bw allocation for different traffic types, or QoS), excessive IP traffic might cause problems to VoIP calls too. And that would include virus-generated traffic amongst 'valid' overloads.
Of course this doesn't (like the original author claimed) make VoIP impractical or too suspectible to fluctuations on general Internet traffic rates, but perhaps it could cause some problems. But... that's something companies have to deal with whether viruses exist or not.
--
They are not talking about using stock linux (kernel) as is. However, linux has been used extensively in embedded world, and customized (light-weight) versions may well make sense for many mobile phones. Mobile phones need an OS, and having that OS be based on something widely used in other places (from other embedded systems to mainframes) does have its own benefits, like availability of mature tools (gcc, debuggers etc.).
... and I've heard there are even people using Windows/b (CE) on tiny devices... --
Operating System traditionally (before Microsoft "redefined" the term) meant system software that abstracts hardware to applications (or at least defines interfaces to device drivers outside OS), handles access to shared resources like memory, and does some level of process management. Thus, except for the most primitive devices, all electronic devices usually do have OS, for good reasons.
If you are old enough to have used C-64 (or other home computers of 80s), consider the fact that even C-64 had an OS.:-)
Whether OS should be visible to end user is completely different issue of course. Usually user doesn't (and needn't) care about which OS is in there, Symbian, QNX, Linux or some home-grown mini-OS (like the ones Nokia and Ericsson used to have). But OS is there; it makes no sense to write all the h/w management stuff from scratch for each new device.
--
In any case it is kind of cool if IE is doing that. Not reinventing the wheel I mean. Makes me wonder why Netscape can't use Quicktime on Mac as well?
ONe thing I forgot to mention, though, is that Mozilla (thank god) renders PNGs pretty decently as well, both on Mac, Windows and Linux. It would be a shame if Open Source application didn't support 'Open Format'.
Hehe. Nice idea... But how well would it go with the "paperless office" ideas? About the first idea... umh... having work stations with Internet access in restrooms sounds kind of kinky.
It would be a definite plus to have a catchy name but weird names haven't really stopped other formats from becoming de facto standards. I don't think Linux, BSD, Gimp, gcc, .wav, .jp(e)g or dozens of other names/abbreviations had high coolness factor (ok some of those are still 'geek-only')
Hmmh. Wonder how much better would Linux have done if it had been named "Penguin"? ("Just Pengo for friends!") 8-)
So... I for one have been encoding in ogg for past 3 months (and will continue doing so), and if enough people do the same, ogg will become popular enough, more apps will support it, and level of top ogg-aware applications will rise to the level of top mp3-aware applications (most likely as those will then be the same apps...).
It would be interesting to know, though, how will ogg do in 'non-mp3' areas. I understood that it was at some point sponsored by internet-broadcasting company, ie. it was (also) meant to be a good streaming audio format. Becoming popular (alternative for RealAudio?) there shouldn't hurt its popularity in more 'static' use either.
IE 5 for Mac most certainly does show PNGs. And what is puzzling is that it has by far the best PNG support (better than IE on Windows). I'm not a Mac-user, but I did develop PNG export functionality to a large shrink-wrapped application, and had to test it on various platforms. IE 5/Mac was surprisingly good, even doing correct gamma correction.
Also, if I recall correctly, NS 4.7x did show PNGs on Mac, albeit as badly as on any other platform (I don't have a mac to check this on right now). :-/
It is a pity in any case that NS fails miserably with transparency (except for 8-bit simple transparency images). IE on Windows is slightly better, but can't do 24-bit full alpha either.
You have valid points about why it's not all that likely Palm would be buyer. However, I think moderators thought it was genuinely interesting, even if not realistic or very insightful? And no, obviously I wasn't moderating it. :-)
100-to-1 against Sun. Although they hate the Windows monopoly, they've raised their hands there... And concentrate on server side, enterprise computing etc. They have the money, but there's probably no place for BeOS in their strategy (unfortunately...). Would have been interesting to know what-could-be - case if they had bought Commodore (meaning Amiga) way back when (which they were considering).
Anyways; I'd say that even with trouble, go with an external ISP (instead of Qwest.net). It is worth it.
People should perhaps consider using separate company for their physical line (DSL) and actual Internet services (ISP). Combining those might produce small savings (for me about 5$... but no static IP, just 2 email addresses, no servers allowed etc. etc.), but it might not be worth the hassle. Since Qwest is cool enough (or non-monopolistic... as unbelievable as it is for ex-US West) to allow 'competing' ISPs, why not use this option? Competition really is the key. Besides, personally I think Qwest should be making much much more profit on providing the low-level connection and letting outside ISP take care of costy end-user help desk. I mean; ISP is the place people first contact; and they only get (in my case) 40% of the cake...
Disclaimer: I don't know if this is possible everywhere Qwest has DSL-service. I know it is in Colorado, though.
However, installing linux on your PC is easy enough that you can get your raw socket access easy enough from home. And people who hack other machines to do spoofing usually get to root if they get any normal user account.
It is futile trying to prevent these kinds of access from single user desk top systems.
But like others have said, the biggest problem really is that many people find installing _any OS_ difficult. And most of them need NOT install anything; PCs usually come with some flavour of Windows installed by default. Default installation of Red Hat (and I'm sure others like Mandrake, S.U.S.E) are really trivially easy (as long as h/w really is 'old enough' to be fully supported), and comparable to Windows ME install (which I had to re-install, thanks to the seller having partitioned 30 gig HD in one partition... ugh). After installing Red Hat (for example), updating system is as easy as with windows, if you use RH's update agent (compared to windowsupdate.com).
--
Part of the alleged problems is that the other company is not obeying the restrictions of (Open/)Free Software license in question (GPL). Since companies are involved, there are more relevant issues (relevant to the law suit(s)) contractual issues but I'm sure you can read about those in the article linked to as well as from previous ./ - articles about the case.
--
But more importantly, if you sell your house for $200.000, and only get $45.000 will you be happy that you got "lots of money" for that? Or will you perhaps be mad at getting screwed? If the deal was for 2.5 M U$ then 300 k U$ really is "only"?
--
It's just that it depends on what do you want more; win the legal cases, or try to stay in terms with developers. Perhaps you can't do both... Just sounds like if I had to do that, I'd feel dirty as hell, bending over and letting lawyers run the show, just like Evil Big Companies do. Even if I knew it may be the smartest thing to do.
--
Well, if MySQL AB employs the actual authors/copyright holders of MySQL (like I think it does?), they can dual-license their code (triple license etc) however they want? Hell, they could create "users have to wear rainbow coloured donkey hats every friday if they use our product" license for NuSphere. :-)
If they were/are using GPL'ed code created by other people/companies, this would of course be different... But I think that's more what NuSphere is doing, and having problems with.
It would be interesting to know how other companies in similar situation (Transvirtual with Kaffe JVM, for example dual-licenses various versions of Kaffe) deal with problems like this? Are there other GPL-using companies that have created useful-enough applications that other companies are creating plug-ins/additions/enhancements?
--
In one of the previous threads related to the ongoing battle, there was an interesting note by someone familiar with history of NuSphere (ex-Progress). Basically what the writer said was that NuSphere most likely had/has problems dealing with Open/Free S/W since it used to be more traditional properietary code company. That probably explains many problems and (potential) misunderstandings.
I still can't blame MySQL AB for trying to make sure game rules are obeyed. It may have come as a surprise to NuSphere what exactly they get and what they don't get (ie. they weren't given a blanket "discard GPL restrictions for your product" rights... those could be given by dual licensing), but like they say, "ignorancy is no excuse"?
--
And this, like he said, is probably even worse for than having to live with full restrictions GPL would impose.
--
Well, I probably should have clarified that I mean security patch - type upgrades. And yes, on work station - side, things get pretty obsolete. But for servers, although (large) companies dislike upgrading to new versions, they do usually apply the patches as required. Or perhaps that's only for the ones that are more mission critical than others, my view may be bit distorted. I'm sure mom and pop - ISPs are different from Fortune 500 dinosaurs.
--
Ah, those lil bastards! Wonder where I could order more of them? :-)
(yes, you got me, my native language has no articles... tricky bastards I'll tell you!)
--
Commercial companies make backup copies of their databases on regular basis. This means that while financially losses are big (big bucks DB admins get paged to restore stuff from backups), the content losses are likely to be at most as severe as results of hardware failures (which do happen... even with sophisticated raid systems etc).
Also, unlike ignorant end users, companies (their admins) usually try to keep their systems up-to-date, and are likely to be less vulnerable to attacks. Even though as targets they are more visible, and probably more lucrative, too, they are much better prepared against threats than your regular Joe Sixpack.
Up until now, we have delt mainly with simple scripts whose workings are obvious.
I don't want to flame you here (you did say you are not a security expert), but usually worms are not just simple scripts (nor even non-word viruses); on unix-systems they may (and have) been scripts to be more portable, but there isn't anything simple in them either.
As to email being required... for decades (since first worms were created, early 80s?) worms have been able to use other network connections than email. That's the case with CR; variety is good for viruses and worms. Spreading using attachments is easy (some might say lame...) way to spread, but bit too obvious. Easy to implement, though, which is why it has been a popular approach.
I guess I just disagree with doomsday prophecies like this. Even though I don't want to appear like a MS-bashing zealot, I must say that Microsoft is now paying for putting security related issues on rather low priority for years. There's a lot that have been done by other companies and organizations (Java-security model by Sun, xBSD code inspections to build reasonably secure server OSes, etc. etc); Microsoft just didn't think potential risks were big enough. They have been proven wrong... and hopefully have started paying more attention.
--
Wonder if sendmail has lost its position as number one Intruder Service, then? :-)
--
Of course this doesn't (like the original author claimed) make VoIP impractical or too suspectible to fluctuations on general Internet traffic rates, but perhaps it could cause some problems. But... that's something companies have to deal with whether viruses exist or not.
--
--
Whether OS should be visible to end user is completely different issue of course. Usually user doesn't (and needn't) care about which OS is in there, Symbian, QNX, Linux or some home-grown mini-OS (like the ones Nokia and Ericsson used to have). But OS is there; it makes no sense to write all the h/w management stuff from scratch for each new device.
--