Client-side error checking is an optional extra. It's not necessary. The only things that are necessary to take information down from people are standard HTML forms that work in any browser.
In particular, a lot of people use client-side error checking instead of server side, for which they should be taken outside and shot. Client side error checking is a nice thing so the user doesn't waste a form submission, but is not a replacement for properly validating user input.
Sorry, just fed up with everyone who knows Visual Basic and/or what the <script> tag does calling themselves a web application developer.
Only time I've seen a OS X kernel panic was due to faulty hardware (mainboard in one, memory in another). Probably worth seeing if you can try some different RAM, at least, see if that helps. Maybe someone else can suggest more useful diagnostics...
Err, did you just move from the US, perhaps? They've always done this. Part of it is tax (at 17.5%), but mostly they're just being evil money grabbing...
*takes a deep breath*
Oh, and that's not just launch prices. 50-55 Euros for a PSP game is around what I'd expect, given the prices of games for other consoles.
Particularly interesting (given that many people in the US seem to be complaining how expensive it is) is that the XBox 360 will be one of the cheapest console releases in the UK, ever. At only 220 UKP for the basic console, it's a lot cheaper than the 300 UKP that the Playstation, Playstation 2 and XBox all launched at.
My advice: If you're comfortable ordering internationally, go with it. This is nothing more than Sony & co. trying to grab as much money as they can.
Kinda with you on this. I'm not a big fan of Cell - I can't shake a feeling it's overhyped - but I do think multi-core/multi-processer systems are the way forward.
In particular, I don't think they're going to have that big a problem porting between platforms. Split your game engine into a generic CPU-orientated thread, plus 6 threads designed to work well with the various cores in Cell. Admittadely, this leaves one non-generic core unused, but I don't think that's going to be an issue for games in the short to medium term, at least.
Port to XBox 360 by putting the generic thread on one cores, and 3 of each of the smaller threads on the other two cores. Port to PC by stuffing all 7 threads on one CPU (or let the system auto-balance if you have multiple cores).
It's a bit of a hassle, and there will have to be platform specific tweaks, but I don't think that's what's really getting to developers. I think they're not used to having to deal with the issues related to multi-threading, and that's what scares them.
Wish I had mod points, but was about to say exactly the same thing. The people that are going to do this sort of thing aren't the sort of people that are going to know it won't work. Or may just figure they can run really fast to the scanner.
> In my personal opinion, the perfect compromise would be HD-DVD with Blu-Ray sized layers and data storage.
I believe everyone involved would like to see that. However, putting as much data onto a disc as Blu-Ray does means significant changes, which is what HD-DVD is trying to avoid. I personally think Blu-Ray will succeed, as while it may be more expensive to manufacture the discs, unless they're dramatically more expensive, or the people backing it decide to shoot themselves in the foot price wise, actual price difference to the consumer is likely to be negligable.
To be honest, I think most Blu-Ray producers will just take the difference out of the profit margin, on the basis that they want to have the format widely used. The fact that the PS3 will play Blu-Ray discs is also likely to pay a massive part in making the format popular...
Actually, no. Vivendi took Valve to court over Steam, as it allowed Valve to distribute their software without the publisher (Vivendi) taking a cut. Blame for Steam lies solely with Valve...
Where from? Pumps on the scale we're talking about aren't exactly lying around, they're manufactured to order. Then there's the problem of power, given the extensive damage. Then we can start talking about working conditions for installing those pumps...
One thing I wonder about, is how many geeks have female _friends_. Not someone you're dating/married to, but women you just spend time around, talking, watching movies, playing computer games, whatever? Thing is, I think a lot of geeks don't spend time around many women. Sure, there's a significant other, and maybe a friend or two, but I think spending time around more gender balanced groups would help.
> I'm still paying for that crap, ten years later. It's totally not worth it.
Tell me about it. I never did extended awake time, but instead spent about 2 years averaging 5 hours sleep a night, including one fortnight where I had something on the order of 2 hours sleep a night.
I look back at this, and particularly moments like falling asleep while talking to someone, and wonder what I was thinking.
That's because they're something which is meant to be like sugar, but are explicitely not sugar. Unless I'm missing something here, this should be like real meat, except without any of the complexity of having to be an animal. I think the biggest risk is that it will lack variety...
Given there's lots of talk about the PS3 being an expensive console, and that it will be worth it because it will have a 10 year lifecycle (I'm assuming that's in the same way the PS one is just hitting it's end of lifecycle), I wouldn't bank on the PS3 coming any cheaper...
We design to the standard. Test primarily in Firefox, then also in Safari. Opera tends to work perfectly with anything that looks good in both Firefox and Safari. We then have to tweak things a bit here and there to get it to look okay in IE, but it's hardly unusable!
In particular, I wish people would stop with this obsession that websites must attract my attention. If I'm at a website, it has my attention, now it should provide what I want, quickly. A terribly designed (bright clashing colours, crazy numbers of frames, the entire page rendered in PRE tags) will tend to make me go find somewhere else that's easier to read, but equally some Flash abomination that I have to wrestle with to get what I want, will drive me to another site.
Huh? As part of a web application development team, our policy is to write to the stanadards, then tweak to make it work in as many browers as possible. Our applications may not look as fancy as some places, but so what? I'm fed up with sites insisting I use flash to access them, for example. I'm not there to drool over graphics, I'm there to get some information, and don't want to have to wrestle the interface to get at it.
Sure, if you start playing with Javascript, you can get into a real mess, but I'd have felt that just means you should avoid Javascript.
In particular, if a blind user comes to our website, they should see something useful. We can't simulate this perfectly, but we do test the applications in Lynx, and make sure they're usable.
Really, what's so hard about making websites that work in standards compliant browsers?
Why would I want to use a server side application where a client side one would be fine? If I want cross-platform portability, I'd write it in Java. If we're talking about clients onto data held on a central server elsewhere, sure makes sense. Or, I could use X11, that not only works, but is actually designed for this sort of thing!
On the other hand, word processors. Sure, I'd LOVE to lose the ability to edit documents if there's a network problem. Oh, and are you storing my files locally or remotely?
Okay, client-side Java isn't perfect. Start-up time is a hassle, for example, and Swing still doesn't look so great. However, if people really wanted cross-platform applications, I think we'd have seen more work on improving both areas.
Stop trying to mangle HTML into every situation, please!
I think you're missing the point of AJAX: it's not about making things faster, it's about adding features such as:
Server-push of data (so if something updates server-side, you can push it out to clients).
No need to refresh the page to commit changes - changes can be sent to the server without the user having to submit the page. This is mostly good because it means the client doesn't have to re-render the entire page. It's also good if your users tend to forget to submit after making changes.
It can also use less bandwidth, but that's more a side-effect than anything else, IMHO.
So it's OK to steal that candy bar from a random stranger (well, company), but teaching them basic cause and effect by turning them in? Hell, I'd practically consider it doing them a favour, before someone turns them in for stealing something more expensive, and they get into serious trouble. Better to end up paying for the candy bar (which, at the end of the day, is all that's going to happen to them), than ending up in jail for something bigger.
Irrespective of anything else, I'd certainly avoid anyone I knew did this (not just heard rumours, knew), and watch my back any time I couldn't avoid them.
Assuming the $60 price tag is even right, I'd be extremely skeptical that's anything except an initially high price, for launch. Xbox games in the UK started off at something ridiculous like $80 each, and then rapidly dropped to be inline with other games (err, about $60. But remember, we're not being ripped off here, goodness no).
Another point to keep in mind that not all your programmers have to the above average, or even average. For example, we have a student helping out over summer. She just doesn't have the breadth of experience for us to trust her with architecture design or the most complex programming tasks, but there are lots of simpler, straight forward jobs that she can do. And working at around half my salary (per hour), she's a hell of a lot cheaper than hiring a graduate programmer with several years experience.
I'm assuming you're referring to the fact that BitTorrent doesn't actually work at all well with TCP's congestion control methods? This is a solvable problem:
Basically, TCP is not the ideal protocol for BitTorrent. It is designed for passing around data which both needs to have a guaranteed arrival, and must arrive in order. The congestion control methods are also applied (typically) on a per connection basis, meaning that having a very large number of connections reduces the effectiveness of its congestion control.
A better solution would be for the files to be transferred over UDP, with control protocols run over TCP. Each UDP packet would contain the offset of its content, in the torrent. Receiving these packets out of order is not a problem, as BitTorrent already handles downloading files in an out-of-order manner. Losing a packet merely means that the client has to remember to ask for that part of the torrent again, later.
Congestion management then needs to be done both across individual UDP connections, and over all transfers from a single client. So if the client notices packets being dropped across all connections, it's likely that it is flooding the uplink, and backs off. On the other hand, if only a single connection is losing packets, it needs only back off the transfer rate on that connection, as it is probably a bottleneck to that specific host.
BitTorrent is a brilliant method of localising bandwidth. For example, if I download the latest popular movie from the US (and I live in the UK), then there has to be sufficient bandwidth between my house, down to the ADSL exchange, across to my ISP, up to their trunk provider, under the ocean, along to the host's ISP, and into their server.
On the other hand, if I can download it from my next door neighbour, all you need is more bandwidth down to the telephone exchange. Couple of miles of high bandwidth cabling, instead of a couple of thousand miles!
I'm not saying companies will pass this along, or that ISPs won't throw a fit at the idea, but this is how it _should_ work
If you read the post carefully, it sounds a lot like they have official secured points, but occaisionally departments will get their own wireless access points and set them out, leaving IT services out of the loop.
It really shouldn't be a massive leap to believe that staff, when told by IT services that they have to secure any wireless access points, will just install their own points "because it's easier". I've certainly seen it happen at where I work.
Yup, you're right. Odd, thought 1080i... oh, wait, it'll be 1080p that was the big selling point of PS3 (can't find any details on it right now, unfortunately).
In particular, a lot of people use client-side error checking instead of server side, for which they should be taken outside and shot. Client side error checking is a nice thing so the user doesn't waste a form submission, but is not a replacement for properly validating user input.
Sorry, just fed up with everyone who knows Visual Basic and/or what the <script> tag does calling themselves a web application developer.
Only time I've seen a OS X kernel panic was due to faulty hardware (mainboard in one, memory in another). Probably worth seeing if you can try some different RAM, at least, see if that helps. Maybe someone else can suggest more useful diagnostics...
Err, did you just move from the US, perhaps? They've always done this. Part of it is tax (at 17.5%), but mostly they're just being evil money grabbing...
*takes a deep breath*
Oh, and that's not just launch prices. 50-55 Euros for a PSP game is around what I'd expect, given the prices of games for other consoles.
Particularly interesting (given that many people in the US seem to be complaining how expensive it is) is that the XBox 360 will be one of the cheapest console releases in the UK, ever. At only 220 UKP for the basic console, it's a lot cheaper than the 300 UKP that the Playstation, Playstation 2 and XBox all launched at.
My advice: If you're comfortable ordering internationally, go with it. This is nothing more than Sony & co. trying to grab as much money as they can.
Kinda with you on this. I'm not a big fan of Cell - I can't shake a feeling it's overhyped - but I do think multi-core/multi-processer systems are the way forward.
In particular, I don't think they're going to have that big a problem porting between platforms. Split your game engine into a generic CPU-orientated thread, plus 6 threads designed to work well with the various cores in Cell. Admittadely, this leaves one non-generic core unused, but I don't think that's going to be an issue for games in the short to medium term, at least.
Port to XBox 360 by putting the generic thread on one cores, and 3 of each of the smaller threads on the other two cores. Port to PC by stuffing all 7 threads on one CPU (or let the system auto-balance if you have multiple cores).
It's a bit of a hassle, and there will have to be platform specific tweaks, but I don't think that's what's really getting to developers. I think they're not used to having to deal with the issues related to multi-threading, and that's what scares them.
Wish I had mod points, but was about to say exactly the same thing. The people that are going to do this sort of thing aren't the sort of people that are going to know it won't work. Or may just figure they can run really fast to the scanner.
Disturbingly, Mircosoft seem the less evil option here...
> In my personal opinion, the perfect compromise would be HD-DVD with Blu-Ray sized layers and data storage.
I believe everyone involved would like to see that. However, putting as much data onto a disc as Blu-Ray does means significant changes, which is what HD-DVD is trying to avoid. I personally think Blu-Ray will succeed, as while it may be more expensive to manufacture the discs, unless they're dramatically more expensive, or the people backing it decide to shoot themselves in the foot price wise, actual price difference to the consumer is likely to be negligable.
To be honest, I think most Blu-Ray producers will just take the difference out of the profit margin, on the basis that they want to have the format widely used. The fact that the PS3 will play Blu-Ray discs is also likely to pay a massive part in making the format popular...
Actually, no. Vivendi took Valve to court over Steam, as it allowed Valve to distribute their software without the publisher (Vivendi) taking a cut. Blame for Steam lies solely with Valve...
Where from? Pumps on the scale we're talking about aren't exactly lying around, they're manufactured to order. Then there's the problem of power, given the extensive damage. Then we can start talking about working conditions for installing those pumps...
One thing I wonder about, is how many geeks have female _friends_. Not someone you're dating/married to, but women you just spend time around, talking, watching movies, playing computer games, whatever? Thing is, I think a lot of geeks don't spend time around many women. Sure, there's a significant other, and maybe a friend or two, but I think spending time around more gender balanced groups would help.
Or I could be completely wrong. Thoughts anyone?
> I'm still paying for that crap, ten years later. It's totally not worth it.
Tell me about it. I never did extended awake time, but instead spent about 2 years averaging 5 hours sleep a night, including one fortnight where I had something on the order of 2 hours sleep a night.
I look back at this, and particularly moments like falling asleep while talking to someone, and wonder what I was thinking.
That's because they're something which is meant to be like sugar, but are explicitely not sugar. Unless I'm missing something here, this should be like real meat, except without any of the complexity of having to be an animal. I think the biggest risk is that it will lack variety...
Given there's lots of talk about the PS3 being an expensive console, and that it will be worth it because it will have a 10 year lifecycle (I'm assuming that's in the same way the PS one is just hitting it's end of lifecycle), I wouldn't bank on the PS3 coming any cheaper...
What?
We design to the standard. Test primarily in Firefox, then also in Safari. Opera tends to work perfectly with anything that looks good in both Firefox and Safari. We then have to tweak things a bit here and there to get it to look okay in IE, but it's hardly unusable!
In particular, I wish people would stop with this obsession that websites must attract my attention. If I'm at a website, it has my attention, now it should provide what I want, quickly. A terribly designed (bright clashing colours, crazy numbers of frames, the entire page rendered in PRE tags) will tend to make me go find somewhere else that's easier to read, but equally some Flash abomination that I have to wrestle with to get what I want, will drive me to another site.
Huh? As part of a web application development team, our policy is to write to the stanadards, then tweak to make it work in as many browers as possible. Our applications may not look as fancy as some places, but so what? I'm fed up with sites insisting I use flash to access them, for example. I'm not there to drool over graphics, I'm there to get some information, and don't want to have to wrestle the interface to get at it.
Sure, if you start playing with Javascript, you can get into a real mess, but I'd have felt that just means you should avoid Javascript.
In particular, if a blind user comes to our website, they should see something useful. We can't simulate this perfectly, but we do test the applications in Lynx, and make sure they're usable.
Really, what's so hard about making websites that work in standards compliant browsers?
Why would I want to use a server side application where a client side one would be fine? If I want cross-platform portability, I'd write it in Java. If we're talking about clients onto data held on a central server elsewhere, sure makes sense. Or, I could use X11, that not only works, but is actually designed for this sort of thing!
On the other hand, word processors. Sure, I'd LOVE to lose the ability to edit documents if there's a network problem. Oh, and are you storing my files locally or remotely?
Okay, client-side Java isn't perfect. Start-up time is a hassle, for example, and Swing still doesn't look so great. However, if people really wanted cross-platform applications, I think we'd have seen more work on improving both areas.
Stop trying to mangle HTML into every situation, please!
I think you're missing the point of AJAX: it's not about making things faster, it's about adding features such as:
Server-push of data (so if something updates server-side, you can push it out to clients).
No need to refresh the page to commit changes - changes can be sent to the server without the user having to submit the page. This is mostly good because it means the client doesn't have to re-render the entire page. It's also good if your users tend to forget to submit after making changes.
It can also use less bandwidth, but that's more a side-effect than anything else, IMHO.
Let me get this straight:
Theft, and assault, are both fine
Teaching you to take responsibility for your own actions is wrong
???
So it's OK to steal that candy bar from a random stranger (well, company), but teaching them basic cause and effect by turning them in? Hell, I'd practically consider it doing them a favour, before someone turns them in for stealing something more expensive, and they get into serious trouble. Better to end up paying for the candy bar (which, at the end of the day, is all that's going to happen to them), than ending up in jail for something bigger.
Irrespective of anything else, I'd certainly avoid anyone I knew did this (not just heard rumours, knew), and watch my back any time I couldn't avoid them.
Assuming the $60 price tag is even right, I'd be extremely skeptical that's anything except an initially high price, for launch. Xbox games in the UK started off at something ridiculous like $80 each, and then rapidly dropped to be inline with other games (err, about $60. But remember, we're not being ripped off here, goodness no).
Another point to keep in mind that not all your programmers have to the above average, or even average. For example, we have a student helping out over summer. She just doesn't have the breadth of experience for us to trust her with architecture design or the most complex programming tasks, but there are lots of simpler, straight forward jobs that she can do. And working at around half my salary (per hour), she's a hell of a lot cheaper than hiring a graduate programmer with several years experience.
I'm assuming you're referring to the fact that BitTorrent doesn't actually work at all well with TCP's congestion control methods? This is a solvable problem:
Basically, TCP is not the ideal protocol for BitTorrent. It is designed for passing around data which both needs to have a guaranteed arrival, and must arrive in order. The congestion control methods are also applied (typically) on a per connection basis, meaning that having a very large number of connections reduces the effectiveness of its congestion control.
A better solution would be for the files to be transferred over UDP, with control protocols run over TCP. Each UDP packet would contain the offset of its content, in the torrent. Receiving these packets out of order is not a problem, as BitTorrent already handles downloading files in an out-of-order manner. Losing a packet merely means that the client has to remember to ask for that part of the torrent again, later.
Congestion management then needs to be done both across individual UDP connections, and over all transfers from a single client. So if the client notices packets being dropped across all connections, it's likely that it is flooding the uplink, and backs off. On the other hand, if only a single connection is losing packets, it needs only back off the transfer rate on that connection, as it is probably a bottleneck to that specific host.
BitTorrent is a brilliant method of localising bandwidth. For example, if I download the latest popular movie from the US (and I live in the UK), then there has to be sufficient bandwidth between my house, down to the ADSL exchange, across to my ISP, up to their trunk provider, under the ocean, along to the host's ISP, and into their server.
On the other hand, if I can download it from my next door neighbour, all you need is more bandwidth down to the telephone exchange. Couple of miles of high bandwidth cabling, instead of a couple of thousand miles!
I'm not saying companies will pass this along, or that ISPs won't throw a fit at the idea, but this is how it _should_ work
If you read the post carefully, it sounds a lot like they have official secured points, but occaisionally departments will get their own wireless access points and set them out, leaving IT services out of the loop.
It really shouldn't be a massive leap to believe that staff, when told by IT services that they have to secure any wireless access points, will just install their own points "because it's easier". I've certainly seen it happen at where I work.
Yup, you're right. Odd, thought 1080i... oh, wait, it'll be 1080p that was the big selling point of PS3 (can't find any details on it right now, unfortunately).