With At&T, I pay $5 for every 2 megs of bandwidth used. If I buy this card, it would cost me $15 just to download all the porn spam I get every morning. Unless the bandwidth costs get reasonable, I won't buy it.
Youch, that's a lot more expensive than it is here (Sweden) IIRC. That said, if you really need the portable connection, there are ways to minimise the bandwidth needed. Filter your mail on the server, use an IMAP client (mail.app on mac, mutt on *nix, pmail on windows all work alright for this) and set it to only download headers automatically, wait till you request the message to get the body, wait till you request attachments to get them... and browse with lynx. Or at least turn off image loading. Do all that and you'll cut your bandwidth usage tremendously.
It's not a phone, that wouldn't be news, I've got a tri-band GPRS phone in my pack. Plug it in via firewire... anyway this isn't a phone, it doesn't do voice, data only, it's a PMCIA card dedicated GPRS modem. And you're right, it's expensive, mind you the nice thing with GPRS is that you don't pay for time connected, just bits transferred, so if you filter your mail on the server and browse with lynx you could keep constantly up to date relatively cheaply. GPRS service in Europe runs around 20 cents a megabyte IIRC, depending on where you are of course... anyone know about the US? it would definately get expensive fast if you don't take steps to keep your bandwidth usage limited.
You're following their line by associating the damage they are doing with so-called theft at all. The ability to play your CD on your computer, for instance, while it's certainly desirable if you want to 'steal' the music is not by any mean only desirable for that purpose. My laptop is my portable do-it-all device, and with my mobile lifestyle it's also the only cd player I own. I'm in the minority on that, but these days it's a pretty sizeable minority, this is not at all an uncommon pattern. If I buy a CD and it won't play on my laptop, then I'm screwed unless I am willing and able to go to some lengths to defeat the copy-prevention mechanisms and rip the sucker. Now, if I were a big-time "music pirate" I would doubtless do just that, and the music would be on Kazaa or maybe flowing from my assembly line if I were really bigtime in hours. But since I'm not, my only option is to return the defective CD and get my money back. If they try to deny that option too, and call me a 'pirate' the results of that are unpredictable but certainly not beneficial to them.
I guess the point is that this sort of crap will hardly slow down the pirates, not the amateurs and certainly not the larger more organised commercial groups that sell unlicensed CDs, it's really only going to interfere with those who are doing nothing illegal, who just want the CD they bought and paid for to work on their laptop or their carstereo.
These companies are doing nothing but pissing in their own cornflakes... the only way this makes any sense at all is if you assume it's a ploy to play for some huge government bailout... that's the only way a company can treat their customers this way without destroying themselves.
Like benchmarking how log it take to crash. PC's got Apple beat by a logshot.
Hahaha right on brother! Did I mention I've been using this TiBook for 3 months now, it's loaded to the gills with alpha software and legacy crap both, and I've yet to see it crash once?
The supposed speed there is mostly marketing hype. Show me an IDE drive with a sustained transfer rate that even comes close to maxing out UDMA/66? You can't, they don't exist.
SCSI will spin 8 disks at once on the same channel, and that's the only way you'll max out the transfer capacity of the channel. IDE only handles 2 drives per channel, so the difference between UDMA/66 and UDMA/100 amounts to nothing other than feeling 'l33t'. Yes, very rarely, when you have drives that are doing burst transfer from their own cache, you might actually use the difference. But the UDMA/66 will catch right up the next millisecond anyway, no one would actually notice it even then.
The short answer is most drives can't send much more than 66mbit right now anyway.
Umm most? Name one that can.
"Faster" EIDE/ATA specs are nothing but marketing hype, there are no drives on the market that can use them anyway, excepting burst transfer from a cache hit which is really going to have no effect since it happens rarely and it's not sustained, so at most you lose a millisecond but gain it right back the next cycle.
It's a little different with SCSI, since you can run several drives simultaneously, but no ones putting multiple hard drives in laptops anyway, for good reasons... weight, heat, power...
You don't have to shout, I can hear you just fine.
Good. At least one way in which you're different from the usual self-proclaimed web designer.
Seriously, you are making my exact point. This is why designers will use relative widths to ensure their content can be rendered nicely in a variety of interfaces.
And, before doing that, think about whether or not there's really any need to use frames to begin with. There almost always is not.
My assertion is simple: the existence or non-existence of a height scrollbar should not change the relative width of the viewframe. The scrollbars belong to the application, and not the content. I don't know any designer or user who expects a scrollbar to cause a reflow of the contents, shortening or lengthening all responsibly stated relative widths by X pixels.
That's the best argument you had, for sure, unfortunately it was not the first or the loudest. Most of them were quite senseless - for instance claiming that this behaviour was inconsistent with all other browsers which is absolute nonsense - I think the only browser I've seen that does what you want on a Mac is Opera. IE for Mac, NS4, etc. all do the same thing as Mozilla.
So this is a much better argument, yes, but it still fails. Yes, I agree that scrollbars are part of the application, not the content. But many of us do expect that scrollbars are only displayed when needed, they behave that way in numerous instances on numerous different platforms, and it seem quite the sensible way for them to behave.
The fact that you're worried so much about a handfull of pixels tells me you want to control the presentation layer, which is exactly what I disagree with.
You are right: designers should expect the width and height to change. This why we have used percentiles to describe relative widths to make sure things flow nicely, regardless of the interface. Having a situation where the width changes on arbitrary changes to height is, IMHO, plain stupid.
But whether you think it's stupid or not, as a web designer, it's quite simply not your concern. It will happen on some systems, not on others, and it doesn't matter one bit either way.
BTW I looked at your test page, in Mozilla and IE on Mac, and frankly I think you've gotten very worked up about nothing at all. Both of them have the behaviour you object to, and I seriously doubt that one user in 10,000 would even notice it, let alone care. And if I were you, I'd have avoided even the (IMOP tiny) potential for annoyance here by not having used frames to begin with.
Anyway, if the history of that bug, and the conversation threads here say anything, it's that this is not one of those cases where anyone is concretely "right" or "wrong". This is a usability issue, and I would challenge the Moz team (or anyone else) to submit this behaviour to a battery of real usability tests. If it was determined that the majority of users and designers don't mind how a good number of existing pages render, then I'd reconsider.
Well that would be quite expensive of course, but if you'll foot the bill I am quite certain I could probably use the facilities here for the testing.
Until then, all we have is anecdotal evidence, but it certainly seems to me that the current behaviour has as many fans as critics, and that the whole issue is so incredibly minor it just seems absurd to me that you seem to feel it's so important.
I agree completely that this is indeed a case where there isn't necessarily an iron-clad 'right' or 'wrong' behaviour of the browser, but there is a definate right behaviour of the designer, which is not to obsess about pixel-level layout issues.
HTML isn't about pixel level control, it's about making content accessible. I haven't seen anything, here or on bugzilla, to even suggest credibly that this is the usability issue you seem to think it is. And the worst thing they could do for Mozilla would be to start 'fixing' things like this - it's bad enough they wasted as much time as they did talking about it. It's not a bug, at worst it's a 'quirk' - and hardly a noticeable one to anyone not fixated on pixel level control it looks like to me.
I cut and pasted the link and read the whole thing. You're wrong, they're right.
In other words it is not GENERALLY acceptable that the page width is
unpredictable e.g. that it depends on the length of the page.
Hello? HTML 101. The page width, and any other physical attributes of the output device, are unknown and unknowable. That's the entire point to the abstraction involved in HTML, as opposed to.pdf or something. You don't even have to have a screen to parse html to, the end user may well be using a reader. The entire point to using HTML is to mark up the content in such a way that the browser can then determine how to best present it.
Seriously, people like you are killing the web, choking it to death with your bullshit 'I'm a designer' attitude and it really pisses me off. People worthy of the title 'designer' in any field know enough to educate themselves about a particular media before they use it, but for some reason 'web designers' seem to almost universally feel that it works the other way around, that the media should adapt itself to their goals. It's like whining that charcoal needs to be fixed because it doesn't allow you to use colours.
You're not the first person that's mentioned this, my TiBook doesn't seem to have it. Any idea what's wrong? I read that it comes standard with OSX, but it's nowhere to be found for me...
For example, another post in this thread claims that "... a 1 GHz G4 is about the same as a 4 GHz P4 in speed", which I suspect is a little optimistic;)
I have to agree, that's very optimistic. That said, it's not wholly untrue - the numbers will back that up or very close on certain tests - the ones where the Mac does best, primarily involving Adobe Photoshop. That is probably the one application that drives more Mac sales than any others, and it's naturally totally optimised for the Mac architecture as a result.
On other benchmarks they don't fare quite so well. I remember a fairly recent SPEC showing where the G4 came in only something like 20% above a PIII at the same clock speed (remember that the PIV benchmarks below the PIII at the same clockspeed) when running against Windows, and dead even when running against Linux. Which doesn't look so good. But this test doesn't use any of the special capabilities of the G4 that optimised Mac applicatons do access... it's probably fairly accurate in reflecting how fast it will run apps that aren't particularly in need of computational power will run, but the ones that do need the power tend to be optimised more than that. Perhaps to the degree that these links would imply, or at least close. GCC is sadly not very good at generating fast G4 code, though it's improving, and that benchmark really tests the compiler moreso than the hardware.
These benchmarks on the Xserves are much more impressive. Really, when it comes to the benchmarks on these things, sometimes they look really good, sometimes they look really crappy. It's definately a mixed bag. But in the end, what's important isn't how it looks on a benchmark but how it performs for you day to day, on whatever it is that you use the computer for. On normal everyday tasks, my TiBook keeps up with Wintel books at twice the clock speed, is lighter, and a lot nicer to work on. That's what counts. YMMV, of course... if you're really curious about the performance go down to the nearest store that carries them and try one out.
Yeah, I think you're right, 5 hours is pretty optimistic. My TiBook (with the G4 of the beast, 666Mhz;) usually gives me about 3 hours, of course I am very hard on batteries, I always get a lot less than quoted. That said, 3 hours is pretty damn good, and if it's not enough an extra battery sets you back only $129.
You need the DRM kernel module. Instructions are here - this is mostly aimed at Debian users but it works for MandrakePPC and YellowDog almost the same way.
My TiBook has 'only' 256mb ram and I've been wishing I'd insisted on 512 - it works fine as is really, but the thrashing when I try to run too many apps, including one BIG one that still runs only under classic, is a little annoying. I hadn't looked into it, good to know it's not a hard upgrade to do yourself, now the question - do these things take regular DIMMS or do you have to buy some sort of special Apple memory?
Don't fall into the Mhz myth, the clock speeds on these things are lower, but they get more work done in a clock cycle too.
That said, they're still a little slower in terms of work done per second than the fastest Intel has, just not nearly as much so as you may think looking at the clock speeds. But it doesn't really matter all that much. CPU speed is just one factor of overall performance, and with a good design it doesn't need to be nearly as fast as it would on a poorer design. The design on the Macs really is much better, the bottlenecks aren't as bad, and they have plenty of power where it counts. Think of it as finesse vs. brawn in comparison with your typical Intel/AMD system, where the surfeit of CPU speed is used to overcome a basically less efficient design. Consider that probably over 90% of the computation done on a pc these days is concentrated in the graphics rendering, and the look at how much more efficiently the mac handles that - all the way from a GPU which is faster than the CPU to the Altivec system in the CPU, which beats the hell out of MMX/SSE and all that.
I'm typing this on a TiBook now, a 666 Mhz G4, and believe me, when I put it up against a new Intel based notebook it won't take the speed crown, by any means, but it's close enough that I don't really care. It will outperform Intel notebooks with over twice the clockspeed quite handily on most tasks, and when you look at things like the screen and the cd-rw/dvd drive... Apple was overpriced once but it's changed. You'd be very hard pressed to find an intel notebook with the same features in the same weight-class much cheaper. And OS X beats WinXP in nearly every category for my money. Easier to use, prettier, AND more powerful under the good as well... tcsh or bash beats cmd.exe any day.
The majority of the processing needed in modern pcs is in fact for all the graphics. So it makes perfect sense to have a faster GPU than CPU - that's where the horsepower is needed. Even if you're doing relatively computationally intense work (I run statistical analyses daily) the requirement still don't add up to the level required to run Aqua or WinXPs graphics.
Remember the Amigas? Positively legendary machines, and for good reason, they were designed this way. The CPU wasn't much at all by modern standards, but it was enough to do what it needed to do just fine (and, in all honesty, enough to handle the non-graphical computations done on most pcs to this day.) The Video Toaster was capable of working pretty much independant of the CPU, and it had a lot more horsepower... the end result was a machine that surpassed PCs made many years later in real functional power.
Re:Kazaa vs. eDonkey
on
Gnutella2?
·
· Score: 2, Interesting
Wow, that's a big change. I wonder where that leaves the security issue though, that's of course always been coming, but now I guess it's upon us... the network relies on the fact that even those who try to be leechers can't avoid sharing the parts already downloaded while waiting for the rest... if the complete source is out it will be much easier for someone to put together a full leecher client... and if that becomes very popular the whole network will become untenable.:(
I never thought security through obscurity was a viable philosophy longterm, but it's better than nothing. What now? Have any of the developers addressed this that you know of?
Re:Kazaa vs. eDonkey
on
Gnutella2?
·
· Score: 5, Informative
If you have a *nix box (even an apple if it's OS X) you can use mldonkey which is a very nice client. You can operate it remotely from another box, it uses both edonkey and overnet protocols simultaneously, it's partially open source (there is a key component kept secret for security reasons, the one flaw in thes protocols is that they require trusted clients unfortunately) and it really gives you the best of edonkey and overnet both, as well as supporting the move to overnet since anything you're downloading from edonkey or sharing out will also be shared to overnet.
There's about 200 of them all told. This is a HUGE company. The major subdivisions (just in the record division, they own a lot in other areas too) are Arista, J Records, RCA Music Group, and BMG Asia, Latin, Europe, Music Publishing, and Distribution. I couldn't find a thorough listing of their 200+ front labels, but I think that if you look at the fine print they'll all say something about being associated with Arista, J Records, RCA, or BMG something or other...
Unlike most of the posters trying to give reasons not to Free the drivers, you've actually made a point that stands up. But there is something odd about it. It only makes sense for a very short time period. Binary drivers compared to source available won't stop the competition from reverse engineering your stuff, it just makes it a bit harder. Within probably 3 months in most markets (and I think that estimate is very much on the high side) the whole issue will have become totally irrelevant. But your customers are then stuck - even though you're gaining no further advantage by keeping the source to yourself, the customers difficulties caused by keeping the drivers closed continue as long as the hardware is still in use, which can easily be years.
So a suggestion to manufacturers in this situation, go ahead and release the binary drivers, a la Nvidia, initially. But, a few months later, when this is no longer the cutting edge hardware of the moment, open the source up. You save yourself a lot of support problems, your customers will love you for it, and you're really not losing anything at that point anyway - the secrecy has already outlived its purpose and gained you everything it could gain you already.
I could see them making the OS gratis. It would really be a smart move on their part, and I really hope they aren't that smart. Because you know, even if they do give it away, it will be free as in gratis but NEVER Free as in liberty. Unfortunately, if they did that, they might distract and confuse enough people to stay in power. And that would be a great tragedy.
Oh yes, there are lots of very intelligent people in very poor countries. There is probably a social-evolutionary element to this - it's easier for the stupid to get on in a rich, well developed country than a place like Namibia. Some of the smartest people I've known came from some of the poorest places.
BTW, not that it matters to your point, but I find it amusing and if you don't have any experience with Africans you may not be aware - DO NOT assume that because the man writes well you'd have no difficulty communicating with him face to face. Places like Namibia, Tanzania, and Nigeria have some very smart people that speak English natively, but their accent is truly unbelievable. If you met this guy in person you'd probably swear he was speaking another language. After being around him for a week or two your ears would adjust to where you could understand him half the time, maybe. It's a beautiful accent really, but it's certainly a shock the first time you run into it. Makes Indian and Pakistani English sound like the girl next door in comparison.
Frankly, this is a pretty bad way to get your name out - an IT company that doesn't understand the web any better than this? I wouldn't hire them to do anything, they sound totally incompetent. But they say any publicity is good publicity...
Slashdot Mirror
Youch, that's a lot more expensive than it is here (Sweden) IIRC. That said, if you really need the portable connection, there are ways to minimise the bandwidth needed. Filter your mail on the server, use an IMAP client (mail.app on mac, mutt on *nix, pmail on windows all work alright for this) and set it to only download headers automatically, wait till you request the message to get the body, wait till you request attachments to get them... and browse with lynx. Or at least turn off image loading. Do all that and you'll cut your bandwidth usage tremendously.
It's not a phone, that wouldn't be news, I've got a tri-band GPRS phone in my pack. Plug it in via firewire... anyway this isn't a phone, it doesn't do voice, data only, it's a PMCIA card dedicated GPRS modem. And you're right, it's expensive, mind you the nice thing with GPRS is that you don't pay for time connected, just bits transferred, so if you filter your mail on the server and browse with lynx you could keep constantly up to date relatively cheaply. GPRS service in Europe runs around 20 cents a megabyte IIRC, depending on where you are of course... anyone know about the US? it would definately get expensive fast if you don't take steps to keep your bandwidth usage limited.
You're following their line by associating the damage they are doing with so-called theft at all. The ability to play your CD on your computer, for instance, while it's certainly desirable if you want to 'steal' the music is not by any mean only desirable for that purpose. My laptop is my portable do-it-all device, and with my mobile lifestyle it's also the only cd player I own. I'm in the minority on that, but these days it's a pretty sizeable minority, this is not at all an uncommon pattern. If I buy a CD and it won't play on my laptop, then I'm screwed unless I am willing and able to go to some lengths to defeat the copy-prevention mechanisms and rip the sucker. Now, if I were a big-time "music pirate" I would doubtless do just that, and the music would be on Kazaa or maybe flowing from my assembly line if I were really bigtime in hours. But since I'm not, my only option is to return the defective CD and get my money back. If they try to deny that option too, and call me a 'pirate' the results of that are unpredictable but certainly not beneficial to them.
I guess the point is that this sort of crap will hardly slow down the pirates, not the amateurs and certainly not the larger more organised commercial groups that sell unlicensed CDs, it's really only going to interfere with those who are doing nothing illegal, who just want the CD they bought and paid for to work on their laptop or their carstereo.
These companies are doing nothing but pissing in their own cornflakes... the only way this makes any sense at all is if you assume it's a ploy to play for some huge government bailout... that's the only way a company can treat their customers this way without destroying themselves.
Hahaha right on brother! Did I mention I've been using this TiBook for 3 months now, it's loaded to the gills with alpha software and legacy crap both, and I've yet to see it crash once?
The supposed speed there is mostly marketing hype. Show me an IDE drive with a sustained transfer rate that even comes close to maxing out UDMA/66? You can't, they don't exist.
SCSI will spin 8 disks at once on the same channel, and that's the only way you'll max out the transfer capacity of the channel. IDE only handles 2 drives per channel, so the difference between UDMA/66 and UDMA/100 amounts to nothing other than feeling 'l33t'. Yes, very rarely, when you have drives that are doing burst transfer from their own cache, you might actually use the difference. But the UDMA/66 will catch right up the next millisecond anyway, no one would actually notice it even then.
Umm most? Name one that can.
"Faster" EIDE/ATA specs are nothing but marketing hype, there are no drives on the market that can use them anyway, excepting burst transfer from a cache hit which is really going to have no effect since it happens rarely and it's not sustained, so at most you lose a millisecond but gain it right back the next cycle.
It's a little different with SCSI, since you can run several drives simultaneously, but no ones putting multiple hard drives in laptops anyway, for good reasons... weight, heat, power...
Gentoo has a PPC port now?
Good. At least one way in which you're different from the usual self-proclaimed web designer.
And, before doing that, think about whether or not there's really any need to use frames to begin with. There almost always is not.
That's the best argument you had, for sure, unfortunately it was not the first or the loudest. Most of them were quite senseless - for instance claiming that this behaviour was inconsistent with all other browsers which is absolute nonsense - I think the only browser I've seen that does what you want on a Mac is Opera. IE for Mac, NS4, etc. all do the same thing as Mozilla.
So this is a much better argument, yes, but it still fails. Yes, I agree that scrollbars are part of the application, not the content. But many of us do expect that scrollbars are only displayed when needed, they behave that way in numerous instances on numerous different platforms, and it seem quite the sensible way for them to behave.
The fact that you're worried so much about a handfull of pixels tells me you want to control the presentation layer, which is exactly what I disagree with.
But whether you think it's stupid or not, as a web designer, it's quite simply not your concern. It will happen on some systems, not on others, and it doesn't matter one bit either way.
BTW I looked at your test page, in Mozilla and IE on Mac, and frankly I think you've gotten very worked up about nothing at all. Both of them have the behaviour you object to, and I seriously doubt that one user in 10,000 would even notice it, let alone care. And if I were you, I'd have avoided even the (IMOP tiny) potential for annoyance here by not having used frames to begin with.
Well that would be quite expensive of course, but if you'll foot the bill I am quite certain I could probably use the facilities here for the testing.
Until then, all we have is anecdotal evidence, but it certainly seems to me that the current behaviour has as many fans as critics, and that the whole issue is so incredibly minor it just seems absurd to me that you seem to feel it's so important.
I agree completely that this is indeed a case where there isn't necessarily an iron-clad 'right' or 'wrong' behaviour of the browser, but there is a definate right behaviour of the designer, which is not to obsess about pixel-level layout issues.
HTML isn't about pixel level control, it's about making content accessible. I haven't seen anything, here or on bugzilla, to even suggest credibly that this is the usability issue you seem to think it is. And the worst thing they could do for Mozilla would be to start 'fixing' things like this - it's bad enough they wasted as much time as they did talking about it. It's not a bug, at worst it's a 'quirk' - and hardly a noticeable one to anyone not fixated on pixel level control it looks like to me.
I cut and pasted the link and read the whole thing. You're wrong, they're right.
Hello? HTML 101. The page width, and any other physical attributes of the output device, are unknown and unknowable. That's the entire point to the abstraction involved in HTML, as opposed to .pdf or something. You don't even have to have a screen to parse html to, the end user may well be using a reader. The entire point to using HTML is to mark up the content in such a way that the browser can then determine how to best present it.
Seriously, people like you are killing the web, choking it to death with your bullshit 'I'm a designer' attitude and it really pisses me off. People worthy of the title 'designer' in any field know enough to educate themselves about a particular media before they use it, but for some reason 'web designers' seem to almost universally feel that it works the other way around, that the media should adapt itself to their goals. It's like whining that charcoal needs to be fixed because it doesn't allow you to use colours.
You're not the first person that's mentioned this, my TiBook doesn't seem to have it. Any idea what's wrong? I read that it comes standard with OSX, but it's nowhere to be found for me...
I have to agree, that's very optimistic. That said, it's not wholly untrue - the numbers will back that up or very close on certain tests - the ones where the Mac does best, primarily involving Adobe Photoshop. That is probably the one application that drives more Mac sales than any others, and it's naturally totally optimised for the Mac architecture as a result.
On other benchmarks they don't fare quite so well. I remember a fairly recent SPEC showing where the G4 came in only something like 20% above a PIII at the same clock speed (remember that the PIV benchmarks below the PIII at the same clockspeed) when running against Windows, and dead even when running against Linux. Which doesn't look so good. But this test doesn't use any of the special capabilities of the G4 that optimised Mac applicatons do access... it's probably fairly accurate in reflecting how fast it will run apps that aren't particularly in need of computational power will run, but the ones that do need the power tend to be optimised more than that. Perhaps to the degree that these links would imply, or at least close. GCC is sadly not very good at generating fast G4 code, though it's improving, and that benchmark really tests the compiler moreso than the hardware.
These benchmarks on the Xserves are much more impressive. Really, when it comes to the benchmarks on these things, sometimes they look really good, sometimes they look really crappy. It's definately a mixed bag. But in the end, what's important isn't how it looks on a benchmark but how it performs for you day to day, on whatever it is that you use the computer for. On normal everyday tasks, my TiBook keeps up with Wintel books at twice the clock speed, is lighter, and a lot nicer to work on. That's what counts. YMMV, of course... if you're really curious about the performance go down to the nearest store that carries them and try one out.
Yeah, I think you're right, 5 hours is pretty optimistic. My TiBook (with the G4 of the beast, 666Mhz ;) usually gives me about 3 hours, of course I am very hard on batteries, I always get a lot less than quoted. That said, 3 hours is pretty damn good, and if it's not enough an extra battery sets you back only $129.
You need the DRM kernel module. Instructions are here - this is mostly aimed at Debian users but it works for MandrakePPC and YellowDog almost the same way.
My TiBook has 'only' 256mb ram and I've been wishing I'd insisted on 512 - it works fine as is really, but the thrashing when I try to run too many apps, including one BIG one that still runs only under classic, is a little annoying. I hadn't looked into it, good to know it's not a hard upgrade to do yourself, now the question - do these things take regular DIMMS or do you have to buy some sort of special Apple memory?
Don't fall into the Mhz myth, the clock speeds on these things are lower, but they get more work done in a clock cycle too.
That said, they're still a little slower in terms of work done per second than the fastest Intel has, just not nearly as much so as you may think looking at the clock speeds. But it doesn't really matter all that much. CPU speed is just one factor of overall performance, and with a good design it doesn't need to be nearly as fast as it would on a poorer design. The design on the Macs really is much better, the bottlenecks aren't as bad, and they have plenty of power where it counts. Think of it as finesse vs. brawn in comparison with your typical Intel/AMD system, where the surfeit of CPU speed is used to overcome a basically less efficient design. Consider that probably over 90% of the computation done on a pc these days is concentrated in the graphics rendering, and the look at how much more efficiently the mac handles that - all the way from a GPU which is faster than the CPU to the Altivec system in the CPU, which beats the hell out of MMX/SSE and all that.
I'm typing this on a TiBook now, a 666 Mhz G4, and believe me, when I put it up against a new Intel based notebook it won't take the speed crown, by any means, but it's close enough that I don't really care. It will outperform Intel notebooks with over twice the clockspeed quite handily on most tasks, and when you look at things like the screen and the cd-rw/dvd drive... Apple was overpriced once but it's changed. You'd be very hard pressed to find an intel notebook with the same features in the same weight-class much cheaper. And OS X beats WinXP in nearly every category for my money. Easier to use, prettier, AND more powerful under the good as well... tcsh or bash beats cmd.exe any day.
The majority of the processing needed in modern pcs is in fact for all the graphics. So it makes perfect sense to have a faster GPU than CPU - that's where the horsepower is needed. Even if you're doing relatively computationally intense work (I run statistical analyses daily) the requirement still don't add up to the level required to run Aqua or WinXPs graphics.
Remember the Amigas? Positively legendary machines, and for good reason, they were designed this way. The CPU wasn't much at all by modern standards, but it was enough to do what it needed to do just fine (and, in all honesty, enough to handle the non-graphical computations done on most pcs to this day.) The Video Toaster was capable of working pretty much independant of the CPU, and it had a lot more horsepower... the end result was a machine that surpassed PCs made many years later in real functional power.
Wow, that's a big change. I wonder where that leaves the security issue though, that's of course always been coming, but now I guess it's upon us... the network relies on the fact that even those who try to be leechers can't avoid sharing the parts already downloaded while waiting for the rest... if the complete source is out it will be much easier for someone to put together a full leecher client... and if that becomes very popular the whole network will become untenable. :(
I never thought security through obscurity was a viable philosophy longterm, but it's better than nothing. What now? Have any of the developers addressed this that you know of?
If you have a *nix box (even an apple if it's OS X) you can use mldonkey which is a very nice client. You can operate it remotely from another box, it uses both edonkey and overnet protocols simultaneously, it's partially open source (there is a key component kept secret for security reasons, the one flaw in thes protocols is that they require trusted clients unfortunately) and it really gives you the best of edonkey and overnet both, as well as supporting the move to overnet since anything you're downloading from edonkey or sharing out will also be shared to overnet.
There's about 200 of them all told. This is a HUGE company. The major subdivisions (just in the record division, they own a lot in other areas too) are Arista, J Records, RCA Music Group, and BMG Asia, Latin, Europe, Music Publishing, and Distribution. I couldn't find a thorough listing of their 200+ front labels, but I think that if you look at the fine print they'll all say something about being associated with Arista, J Records, RCA, or BMG something or other...
Unlike most of the posters trying to give reasons not to Free the drivers, you've actually made a point that stands up. But there is something odd about it. It only makes sense for a very short time period. Binary drivers compared to source available won't stop the competition from reverse engineering your stuff, it just makes it a bit harder. Within probably 3 months in most markets (and I think that estimate is very much on the high side) the whole issue will have become totally irrelevant. But your customers are then stuck - even though you're gaining no further advantage by keeping the source to yourself, the customers difficulties caused by keeping the drivers closed continue as long as the hardware is still in use, which can easily be years.
So a suggestion to manufacturers in this situation, go ahead and release the binary drivers, a la Nvidia, initially. But, a few months later, when this is no longer the cutting edge hardware of the moment, open the source up. You save yourself a lot of support problems, your customers will love you for it, and you're really not losing anything at that point anyway - the secrecy has already outlived its purpose and gained you everything it could gain you already.
There's an even simpler way:
I could see them making the OS gratis. It would really be a smart move on their part, and I really hope they aren't that smart. Because you know, even if they do give it away, it will be free as in gratis but NEVER Free as in liberty. Unfortunately, if they did that, they might distract and confuse enough people to stay in power. And that would be a great tragedy.
The apple mail client, mentioned in the blurb, works very well with IMAP, that's what impressed me enough that I'm actually using it.
Oh yes, there are lots of very intelligent people in very poor countries. There is probably a social-evolutionary element to this - it's easier for the stupid to get on in a rich, well developed country than a place like Namibia. Some of the smartest people I've known came from some of the poorest places.
BTW, not that it matters to your point, but I find it amusing and if you don't have any experience with Africans you may not be aware - DO NOT assume that because the man writes well you'd have no difficulty communicating with him face to face. Places like Namibia, Tanzania, and Nigeria have some very smart people that speak English natively, but their accent is truly unbelievable. If you met this guy in person you'd probably swear he was speaking another language. After being around him for a week or two your ears would adjust to where you could understand him half the time, maybe. It's a beautiful accent really, but it's certainly a shock the first time you run into it. Makes Indian and Pakistani English sound like the girl next door in comparison.
Frankly, this is a pretty bad way to get your name out - an IT company that doesn't understand the web any better than this? I wouldn't hire them to do anything, they sound totally incompetent. But they say any publicity is good publicity...