I once tried finding the date that the UK switched to daylight savings (one of those seasonal clock+1hr things). No joy. Loads and loads of papers on the influence of daylight savings on the economy, road fatalaties, everything but the date of the next changeover itself.
I now rely on the traditional method of hoping someone else will tell me. Seems to work.
FWIW I think Be did the best possible job under extremely difficult circumstances. I mean, obviously the whole idea of the company was to take a pile of cash and some very damn talented engineers and make the new Apple OS. When it became obvious the Apple's new management were not going to buy the company for $Big, they were essentially screwed.... Competing head on with win9x, or a wide variety of embedded unices depending on which story you believed.
That they've hung on as long as they have, that they ported to x86 in double quick time, can only be a tribute to the company's management and the quality of their code.
So why haven't Sony bought them yet?
Anyway, I've said it before - and I'll probably get flamed - but I still believe it: The absolutely best way forward for open source would be to have separate desktop and server OS's. An open source Be would be just wonderful, and this frees up Linux and *BSD for what they do best.
I notice on the screenshot for the login page two things:
(1) The number of running apps that some users have open, including the Administrator.
(2) A button to shut the machine down.
Does this mean that non root^H^H^H^HAdministrator users can shut down higher privaleged (sp) programs? And services? All this time after the original release of NT (1994?) do Microsoft still not understand multi user OS's?
Programmers might not get the satisfaction of building something useful
Au contraire, Rodney. Exactly the reason I left engineering is that no-one in their right mind was going to give me two million quid to make a fast ferry because some hung-over graduate thought it would have fantastic seakeeping. Computing, OTOH, if I think it could be good, I'll sit down and code it. Man, this is way creative.
Dave
DISCLAIMER: Sometimes you are going to have to make software to an engineering quality.
That means keep your day job while developing your next product.
Hell yes. I am currently getting towards the eventual conclusion of developing a product, and this leads me to think of the one piece of advice I'd give to someone at the start of the road I'm at the end of. Keep your day job. Your employer pays you because you have something they want. Think about this in terms of being a business. You're also going to need an 'in' to various places - your current employer is probably as good a place to start as any other.
Mind you, I hated my employer at the time:)
One more thing: This alleged next product - make sure someone with a shedload of money needs it. Like, they cannot live without it quantities of need.
The end result id natural language programming. You literally tell the computer what you want it to do, and its amazing compiler will produce perfect code.
Yup. Exactly. When you (or anyone else) get good, like really good, at C++ or Java or whatever you will eventually get to a state where writing the code is just not difficult. Like, as easy as speaking. Pretty well all the software engineers I've worked with have got to this state. Then all that remains is to explain to the machine exactly what it is you want it to do. Like, exactly. This is almost immeasurably hard and is the chief cause of failure in software projects. Arguably the only cause of failure.
Natural language programming will, kinda obviosly (IMHO) not get around this fact.
Yes and no. I don't have the EEPROM blowing/erasing gear, most motherboards can't mount an EEPROM as a disk (IIRC). And using IDE flash is just so simple.
IMHO: Disaster. Somebody somewhere has to set up the servers, this is most likely going to have to be an ISP. ISP's hate this shit. They hate NNTP, but have to have it, really. They hate IRC, but have it because someone who works there thinks it's great and insists on keeping it up on some godforsaken P200. It's the mother of all security risks. It eats all their bandwidth when it shits it. In this case it'll attract lawyers faster than a hollywood divorce. Uck.
Strict peer-peer will save you. The ISP's will not be bothered by it because as far as they're concerned it's just more layer 3 traffic. Legally they would have a simple time claiming they had no way of policing it. If you're clever and use some encryption, they won't be able to police it.
Peer discovery and propgating queries are the hard bit. How do I know which machines are within (say) a dozen hops? How do I propogate a search? How does it scale.
Yes, it's hard. But a worthy goal, don't you think?
Well, it won't be an MP3, will it? It'll be some proprietary Napster file format. Most likely your Napster client has a private key and a public key. When you get an MP3 from someone else, your client gives their client the public key, their box uses this key to encode the outgoing.mp3 as a.nap, then it can only be played on your box because only your box has the private key.
Piece of piss. How does it play in an MP3 player? It won't.
The thing that Napster are forgetting is why it is they were so successful: It was a classic example of viral marketing. People used it because it was useful. Then more people used it because other people told them it was great - not because they'd seen a billboard advertising it. If they want to charge for napster, and the downloaded MP3's only work on that one machine - its not going to be useful is it?
So the message to those wishing to ego-pander by having (say) 100 million people using their software is simple: Go.
It has to work on Windows, like it or not, and it has to be simple. It has to be secure, it has to scale, it has to be able to search, and it would be damn handy to the lamer in the street if it could cut collections of MP3's to audio CD.
Obviously it has to be completely decentralised, cos RIAA are going to go nuts.
Go for it. Make it work. And do yourself a favour by not putting the letters 'GN' at the start of it's name.
Before consumers buy it, it has to be in the shops. While it's all well and good to yak about selling it over a website, as a sales channel this basically sucks.
Just the logistics of getting things in the shops blows me away. For, like, half a dozen (and upwards) retail chains in - say - thirty countries you have to explain exactly why they should put your Amiga MorphOS Geek/2 Buzzword-Tech thingy on their valuable shelf space rather than boxes and boxes of playstation 2's.
There is indeed, increasingly proving to be a choice amongst appliance vendors, although less so in the 'really' embedded arena. Perhaps a lot of this is related to the traditionally regarded as superior BSD TCP stack - argue what you like, I'm talking traditionally regarded - as well as the fact that you can, and a lot of vendors do, munch around with the kernel as much as you like and keep it proprietary.
Of course, what usually happens is that the people who can munch around with the kernel are the ones who want to release it back into the community and so quite often it does. Case in point: The netgraph networking subsystem - essentially pluggable kernel mode components for making clever network stunts. This was originally made for part of the Whistle (now IBM) box, and has now become an integral part of the FreeBSD 4.x kernel.
Curt Schacker, vice president of corporate marketing for Wind River Corp
Ah, WindRiver - they would be upset. WindRiver sell a really HUGELY expensive (we are talking 'n' kilodollars per developer backside) POSIX OS / toolset for embedded development called VxWorks. They've managed to keep a commanding but certainly not monopolistic lead on the whole embedded OS scene for really quite a while now and the appearance of Linux that's just as good (debate), has shedloads more mindshare and is far cheaper has had a pretty decimating effect on their bottom line. The linked article is a classic example of FUD. Don't sweat it, they're doing it for you. A man called Curt Schacker is doing it for you, actually.
Bypass the whole thing, use BSD. (climbs into flameproof suit).
We can't possibly consider releasing Mozilla 0.8 until they are back up and running, so that lots of Linux zealots and armchair coders can log on and slag off our hard work in a specially-prepared forum.
Fuck, that's funny. So, go on: 95 posts saying that its bloatware and why can't we have a lighter browser; 47 pointing out the obvious and saying that Netscrape 4.7.2 leaks memory; 22 posts on the subject of IE being better; At least some figting pointlessly over whether the UI stinks, or it's just that we don't understand how important XML is...
We are not talking about a couple of people in their back rooms making water cooling kits, flogging them across the net and suing each other over - I dunno - a proprietary hose clip or something.
This is nVidia who are having their arses kicked. What they've done is pulled a fan apart, copied it, and made their own to save, what, US$0.10 per unit on graphics cards that are US$200 - 300 - upwards? And don't forget these things are a shedload more expensive outside the US.
Tightarses, honestly. Fuck'em. Sue their sorry backsides off.
Slashdot Mirror
Wellington, New Zealand. It's been April 2nd here for twenty fucking hours. Enough. Choose one, make it good, don't do this next year - it blows.
Dave
Does this mean the web will eventually become wholly self referrential and crawl up its' own arse? Oh, hang on.....
Dave
I once tried finding the date that the UK switched to daylight savings (one of those seasonal clock+1hr things). No joy. Loads and loads of papers on the influence of daylight savings on the economy, road fatalaties, everything but the date of the next changeover itself.
I now rely on the traditional method of hoping someone else will tell me. Seems to work.
Dave
FWIW I think Be did the best possible job under extremely difficult circumstances. I mean, obviously the whole idea of the company was to take a pile of cash and some very damn talented engineers and make the new Apple OS. When it became obvious the Apple's new management were not going to buy the company for $Big, they were essentially screwed.... Competing head on with win9x, or a wide variety of embedded unices depending on which story you believed.
That they've hung on as long as they have, that they ported to x86 in double quick time, can only be a tribute to the company's management and the quality of their code.
So why haven't Sony bought them yet?
Anyway, I've said it before - and I'll probably get flamed - but I still believe it: The absolutely best way forward for open source would be to have separate desktop and server OS's. An open source Be would be just wonderful, and this frees up Linux and *BSD for what they do best.
Dave
"Rarely is the question asked: Is our children learning?" -- George W. Bush
All your politician are belong to us.
Dave
Is this the famous slashdot "write the whole post twice" glitch? And did I ever get a confirmation email? NAH!
Dave
I notice on the screenshot for the login page two things:
(1) The number of running apps that some users have open, including the Administrator.
(2) A button to shut the machine down.
Does this mean that non root^H^H^H^HAdministrator users can shut down higher privaleged (sp) programs? And services? All this time after the original release of NT (1994?) do Microsoft still not understand multi user OS's?
Dave
Programmers might not get the satisfaction of building something useful
Au contraire, Rodney. Exactly the reason I left engineering is that no-one in their right mind was going to give me two million quid to make a fast ferry because some hung-over graduate thought it would have fantastic seakeeping. Computing, OTOH, if I think it could be good, I'll sit down and code it. Man, this is way creative.
Dave
DISCLAIMER: Sometimes you are going to have to make software to an engineering quality.
That means keep your day job while developing your next product.
:)
Hell yes. I am currently getting towards the eventual conclusion of developing a product, and this leads me to think of the one piece of advice I'd give to someone at the start of the road I'm at the end of. Keep your day job. Your employer pays you because you have something they want. Think about this in terms of being a business. You're also going to need an 'in' to various places - your current employer is probably as good a place to start as any other.
Mind you, I hated my employer at the time
One more thing: This alleged next product - make sure someone with a shedload of money needs it. Like, they cannot live without it quantities of need.
Enough ranting.
Dave
Natural language isn't good enough to precisely express many problems.
:)
Being a better way to express what I said. And shorter. And ironically proving the point in the process, kinda.
Hmmm, time for the blue pills obviously.
Dave
The end result id natural language programming. You literally tell the computer what you want it to do, and its amazing compiler will produce perfect code.
Yup. Exactly. When you (or anyone else) get good, like really good, at C++ or Java or whatever you will eventually get to a state where writing the code is just not difficult. Like, as easy as speaking. Pretty well all the software engineers I've worked with have got to this state. Then all that remains is to explain to the machine exactly what it is you want it to do. Like, exactly. This is almost immeasurably hard and is the chief cause of failure in software projects. Arguably the only cause of failure.
Natural language programming will, kinda obviosly (IMHO) not get around this fact.
Dave
Please tell me this is a joke.
Yes and no. I don't have the EEPROM blowing/erasing gear, most motherboards can't mount an EEPROM as a disk (IIRC). And using IDE flash is just so simple.
So, yeah, I get done out of $40.
Dave
Correct. So you mount / as read only on the flash disk, and put /tmp in a memory disk. AND TURN OFF SWAPPING!!
Dave
IMHO: Disaster. Somebody somewhere has to set up the servers, this is most likely going to have to be an ISP. ISP's hate this shit. They hate NNTP, but have to have it, really. They hate IRC, but have it because someone who works there thinks it's great and insists on keeping it up on some godforsaken P200. It's the mother of all security risks. It eats all their bandwidth when it shits it. In this case it'll attract lawyers faster than a hollywood divorce. Uck.
Strict peer-peer will save you. The ISP's will not be bothered by it because as far as they're concerned it's just more layer 3 traffic. Legally they would have a simple time claiming they had no way of policing it. If you're clever and use some encryption, they won't be able to police it.
Peer discovery and propgating queries are the hard bit. How do I know which machines are within (say) a dozen hops? How do I propogate a search? How does it scale.
Yes, it's hard. But a worthy goal, don't you think?
Dave
Well, it won't be an MP3, will it? It'll be some proprietary Napster file format. Most likely your Napster client has a private key and a public key. When you get an MP3 from someone else, your client gives their client the public key, their box uses this key to encode the outgoing .mp3 as a .nap, then it can only be played on your box because only your box has the private key.
Piece of piss. How does it play in an MP3 player? It won't.
Dave
I doubt they'll take the time to develop a cryptographically secure system.
They don't need to: There are open source (BSD licence) cryptogtraphy toolkits that work perfectly well. The sword cuts both ways.
Dave
The thing that Napster are forgetting is why it is they were so successful: It was a classic example of viral marketing. People used it because it was useful. Then more people used it because other people told them it was great - not because they'd seen a billboard advertising it. If they want to charge for napster, and the downloaded MP3's only work on that one machine - its not going to be useful is it?
So the message to those wishing to ego-pander by having (say) 100 million people using their software is simple: Go.
It has to work on Windows, like it or not, and it has to be simple. It has to be secure, it has to scale, it has to be able to search, and it would be damn handy to the lamer in the street if it could cut collections of MP3's to audio CD.
Obviously it has to be completely decentralised, cos RIAA are going to go nuts.
Go for it. Make it work. And do yourself a favour by not putting the letters 'GN' at the start of it's name.
Dave
None. It'll be done peer-peer.
Dave
If consumers buy it
Before consumers buy it, it has to be in the shops. While it's all well and good to yak about selling it over a website, as a sales channel this basically sucks.
Just the logistics of getting things in the shops blows me away. For, like, half a dozen (and upwards) retail chains in - say - thirty countries you have to explain exactly why they should put your Amiga MorphOS Geek/2 Buzzword-Tech thingy on their valuable shelf space rather than boxes and boxes of playstation 2's.
Go on, why?
Dave
There's always BSD.
There is indeed, increasingly proving to be a choice amongst appliance vendors, although less so in the 'really' embedded arena. Perhaps a lot of this is related to the traditionally regarded as superior BSD TCP stack - argue what you like, I'm talking traditionally regarded - as well as the fact that you can, and a lot of vendors do, munch around with the kernel as much as you like and keep it proprietary.
Of course, what usually happens is that the people who can munch around with the kernel are the ones who want to release it back into the community and so quite often it does. Case in point: The netgraph networking subsystem - essentially pluggable kernel mode components for making clever network stunts. This was originally made for part of the Whistle (now IBM) box, and has now become an integral part of the FreeBSD 4.x kernel.
Cool, eh?
Dave
Curt Schacker, vice president of corporate marketing for Wind River Corp
Ah, WindRiver - they would be upset. WindRiver sell a really HUGELY expensive (we are talking 'n' kilodollars per developer backside) POSIX OS / toolset for embedded development called VxWorks. They've managed to keep a commanding but certainly not monopolistic lead on the whole embedded OS scene for really quite a while now and the appearance of Linux that's just as good (debate), has shedloads more mindshare and is far cheaper has had a pretty decimating effect on their bottom line. The linked article is a classic example of FUD. Don't sweat it, they're doing it for you. A man called Curt Schacker is doing it for you, actually.
Bypass the whole thing, use BSD. (climbs into flameproof suit).
Dave
We can't possibly consider releasing Mozilla 0.8 until they are back up and running, so that lots of Linux zealots and armchair coders can log on and slag off our hard work in a specially-prepared forum.
Fuck, that's funny. So, go on: 95 posts saying that its bloatware and why can't we have a lighter browser; 47 pointing out the obvious and saying that Netscrape 4.7.2 leaks memory; 22 posts on the subject of IE being better; At least some figting pointlessly over whether the UI stinks, or it's just that we don't understand how important XML is...
Good work, Mozilla dudes.
Dave
Ahhh. Doh! That is, indeed, very stupid.
Presumably ADDA simply don't have enough money to make suing them worthwhile. Next time I'll save some bandwidth for vital Napster traffic.
Dave
We are not talking about a couple of people in their back rooms making water cooling kits, flogging them across the net and suing each other over - I dunno - a proprietary hose clip or something.
This is nVidia who are having their arses kicked. What they've done is pulled a fan apart, copied it, and made their own to save, what, US$0.10 per unit on graphics cards that are US$200 - 300 - upwards? And don't forget these things are a shedload more expensive outside the US.
Tightarses, honestly. Fuck'em. Sue their sorry backsides off.
Dave