MacBooks tend to last a lot longer than Windows machines. I'm not talking about build quality here (although that also helps) but about useful lifetime. Any MacBook or MacBook pro built in the last four years can still run the most recent OS, and is therefore good enough to run any software your university course may want you to use. This definitely isn't true for Windows laptops, and especially not for low-end Windows laptops: you'll have a hard time running Win7 on a two year old budget Dell or Acer.
So a lot of these students could be using the hand-me-downs of their older siblings, parents, whatever.
Python's new way of representing floating point numbers is only a partial solution. The bigger problem is accumulation of errors. To see this: add 10 copies of the number "1.1". Now subtract 11. The result will be tiny, but non-zero (-1.8e-15, on my machine).
This is a much more serious problem than the 10.99999999999999998 representation problem.
My feeling is that the basic problem is that, in open source, at most 5% of the people involved are non-programmers (read: non-geeks). And for most projects the number is probably exactly 0% of the people involved. for shareware projects it's close to 50% (half of the developer:-). For commercial projects it's somewhere in the range of 20% (small vendors) to 99% (Microsoft, big software houses).
The input of the non-geeks, while usually dismissed by us geeks as fluff, can be really, really important. Because their interested in such technical trivialities as documentation, ease of use, learning curves, market acceptance (and, yes, financial bottom line too). Those trivialities are important even to hardcore geeks when the software in question is just a tool you need to get the job done (as opposed to the labour of love you've been spending years of your life on).
This article is pretty interesting from a philosophical point of view. One of the first things I realised is that if there is indeed an effect from our measurements it must mean that no sentient beings have been doing such measurements earlier in the lifetime of the universe.
I disagree that the design is "completely irrelevant" for developers. I have three machines in my workspace: 2 macs (PPC and Intel) and one generic PC (for Linux and Windows). The two macs together make less noise than the one PC. Moreover, with two towers under my desk the room there is getting rather limited, so if the third machine hadn't been an iMac I would have had to throw one machine out.
Having three machines may be rather rare, but even with one machine it is really nice if it has a low noise level and a small footprint. It is indeed much more of an issue at home, but in the office it's definitely relevant too.
The 512 MB is enough (but only just so) if you do the average mail/browser/office thing. But on the Intel mac it is *very* enticing to run Windows or Minix alongside MacOSX, using Parallels. Now you suddenly have two operating systems that want to have their working set in memory at all times.
To run Parallels you need at least 1GB, preferably 2. With 512MB you need to pretty much quit all your mac apps before starting Windows, and then there's little advantage over Boot Camp any more...
I'm not too knowledgeable about Theora, but what excites me about Dirac is that it is a wavelet codec. That's exciting, because while wavelet codecs have been studied by computer scientists and (mainly) mathematicians for quite some years this is the first time that there's actually a codec out there that appears to take less than half a day to encode a single frame.
Wavelet codecs could result in an order of magnitude better compression (think: ten times smaller files for the same quality), but Dirac is the first project that I'm aware of that actually tries to test wavelet video compression in the wild.
And the results are quite stunning: much better quality than H.264 at lower bitrates. And because wavelet codes don't work in blocks (like all DCT-related codecs, basically everything else done the last couple of years) any artifacts won't show as ugly square blocks but more as vague smudges, which I find less bothering.
It would be a shame, really, if this would mean the BBC would stop with their efforts to create a state-of-the-art open video codec, just because Real caved in and it was convenient for them to stick with RealVideo....
The BBC Dirac project (dirac.sourceforge.net) was specifically started because of the sorry state of closed codecs, and is technically way ahead of the pack (think: better quality and smaller files than H264 at lower decompression CPU usage), and they're making sure that the format will always be open through very smart use of patents.
I think the real point here, which both Andy and Linus hint on but don't state explicitly (as far as I'm aware) is about keeping the OS designers and implementers honest. If you need an interface between two parts of the system you should design that interface, define it rigidly, then implement it.
Andy likes microkernels because they force you to do that. Time spent on design leads to insight, which may well point to better and cleaner ways to do the task you originally set out to acomplish.
Linus hates microkernels because they force you to do that. Time spent on design is time lost getting working code out the door, and working code will give you experience that will point to better and cleaner ways to do the task you originally set out to acomplish.
There's another advantage to a 25% marketshare for Apple: if after all these years there was serious competition to Microsoft may they'd finally get their act together.
The comments to this article (so far) IMHO show why Open Source user interfaces are in such a bad shape: 90% is about some minor functionality that this-or-that package doesn't have, 9% is about graphics design. Only one post discusses the reason this submission won the contest: it proposes an innovative way to present your daily work.
After 20+ years of research results that tell people what good user interface guidelines are, plus companies such as Apple that have products that more-or-less adhere to these guidelines, it seems that the open source community (I know, equating/. posters with the open source community is a bit of a stretch:-) still doesn't get the point. It is not about how many thousand things your application can do, it is not about beautiful screen layouts, it is about enabling the end user to complete the task they have set themselves with the minimal amount of hassle (especially if s/he has done a similar thing many times before), and helping them with that task as much as possible (especially if s/he is doing something for the first time).
The Nokia 770 could be a very interesting device, I'm anxiously waiting to get my hands on one.
At the moment Linux for organisers is a joke: the linux bit is reasonable but the useability of devices like the Zaurus as a PDA leaves a lot to be desired. And the UI is also pretty abysmal (if you come from a Palm/Macintosh background).
From the whitepapers I've seen from Nokia they are at least putting serious thought into these issues, so we may end up with a device that not only runs Linux but it also actually useable, yeah!
Now we only have to hope that the rumours of the lousy performance turn out to be an issue of the devices still being in beta at the moment...
Slashdot Mirror
MacBooks tend to last a lot longer than Windows machines. I'm not talking about build quality here (although that also helps) but about useful lifetime. Any MacBook or MacBook pro built in the last four years can still run the most recent OS, and is therefore good enough to run any software your university course may want you to use. This definitely isn't true for Windows laptops, and especially not for low-end Windows laptops: you'll have a hard time running Win7 on a two year old budget Dell or Acer. So a lot of these students could be using the hand-me-downs of their older siblings, parents, whatever.
Python's new way of representing floating point numbers is only a partial solution. The bigger problem is accumulation of errors. To see this: add 10 copies of the number "1.1". Now subtract 11. The result will be tiny, but non-zero (-1.8e-15, on my machine). This is a much more serious problem than the 10.99999999999999998 representation problem.
This is almost, but not quite, what I want. I want sloppy focus for mouse-wheel scrolling only.
Is anyone aware of such a thing? It would make my windows work just a wee bit less bothersome...
I gave an impromptu talk at an EuroFOO conference 5 years ago about exactly this problem: http://homepages.cwi.nl/~jack/presentations/OpenSource-EuroFoo.pdf.
My feeling is that the basic problem is that, in open source, at most 5% of the people involved are non-programmers (read: non-geeks). And for most projects the number is probably exactly 0% of the people involved. for shareware projects it's close to 50% (half of the developer:-). For commercial projects it's somewhere in the range of 20% (small vendors) to 99% (Microsoft, big software houses).
The input of the non-geeks, while usually dismissed by us geeks as fluff, can be really, really important. Because their interested in such technical trivialities as documentation, ease of use, learning curves, market acceptance (and, yes, financial bottom line too). Those trivialities are important even to hardcore geeks when the software in question is just a tool you need to get the job done (as opposed to the labour of love you've been spending years of your life on).
This article is pretty interesting from a philosophical point of view. One of the first things I realised is that if there is indeed an effect from our measurements it must mean that no sentient beings have been doing such measurements earlier in the lifetime of the universe.
Having three machines may be rather rare, but even with one machine it is really nice if it has a low noise level and a small footprint. It is indeed much more of an issue at home, but in the office it's definitely relevant too.
The 512 MB is enough (but only just so) if you do the average mail/browser/office thing. But on the Intel mac it is *very* enticing to run Windows or Minix alongside MacOSX, using Parallels. Now you suddenly have two operating systems that want to have their working set in memory at all times. To run Parallels you need at least 1GB, preferably 2. With 512MB you need to pretty much quit all your mac apps before starting Windows, and then there's little advantage over Boot Camp any more...
Wavelet codecs could result in an order of magnitude better compression (think: ten times smaller files for the same quality), but Dirac is the first project that I'm aware of that actually tries to test wavelet video compression in the wild.
And the results are quite stunning: much better quality than H.264 at lower bitrates. And because wavelet codes don't work in blocks (like all DCT-related codecs, basically everything else done the last couple of years) any artifacts won't show as ugly square blocks but more as vague smudges, which I find less bothering.
It would be a shame, really, if this would mean the BBC would stop with their efforts to create a state-of-the-art open video codec, just because Real caved in and it was convenient for them to stick with RealVideo....
The BBC Dirac project (dirac.sourceforge.net) was specifically started because of the sorry state of closed codecs, and is technically way ahead of the pack (think: better quality and smaller files than H264 at lower decompression CPU usage), and they're making sure that the format will always be open through very smart use of patents.
Andy likes microkernels because they force you to do that. Time spent on design leads to insight, which may well point to better and cleaner ways to do the task you originally set out to acomplish.
Linus hates microkernels because they force you to do that. Time spent on design is time lost getting working code out the door, and working code will give you experience that will point to better and cleaner ways to do the task you originally set out to acomplish.
There's another advantage to a 25% marketshare for Apple: if after all these years there was serious competition to Microsoft may they'd finally get their act together.
The comments to this article (so far) IMHO show why Open Source user interfaces are in such a bad shape: 90% is about some minor functionality that this-or-that package doesn't have, 9% is about graphics design. Only one post discusses the reason this submission won the contest: it proposes an innovative way to present your daily work.
/. posters with the open source community is a bit of a stretch:-) still doesn't get the point. It is not about how many thousand things your application can do, it is not about beautiful screen layouts, it is about enabling the end user to complete the task they have set themselves with the minimal amount of hassle (especially if s/he has done a similar thing many times before), and helping them with that task as much as possible (especially if s/he is doing something for the first time).
After 20+ years of research results that tell people what good user interface guidelines are, plus companies such as Apple that have products that more-or-less adhere to these guidelines, it seems that the open source community (I know, equating
From the whitepapers I've seen from Nokia they are at least putting serious thought into these issues, so we may end up with a device that not only runs Linux but it also actually useable, yeah! Now we only have to hope that the rumours of the lousy performance turn out to be an issue of the devices still being in beta at the moment...