To top it off, it really proves the point to see that Little Billy is drooling in the interview about one of the 'up and coming' technologies He is working on - Speech Recognition!
To give him credit, in the interview, he said something to the effect of "the up and coming technologies we're involved with" in reference to speech recognition.
His complete failure to answer any questions in that interview pissed me off, though. Every single question was answered with an advert for Microsoft, whether it was related to Microsoft or not.
And Paxman didn't badger him! I expected him to nail Billy-boy down to an answer!
Sometimes Paxman does wimp out.. he's better at scaring politicians. We'll see what he does on Sunday then.
There was a good Jon Snow interview with Billy-Boy a few months back. Snow did a pretty nice job of making Bill out to be the devil. Let's see how well Paxman does it.
I just have a big telly instead.. even with the black bars, it's still bigger than most widescreen TVs, and is a lot cheaper (although okay, it's lower total resolution) With digital TV and DVD we get the choice of pan-n-scan or letterbox, so I don't think it really matters. Until the widescreen sets are at price parity with 4:3, it won't really take off.
Of course, we don't necessarily want a TV with the clarity of a VGA monitor... the fuzziness hides the pixelisation, especially with those crappy low-bandwidth MPEG streams from digital terrestrial.
I think that's the key to it... as soon as there's a deadline, compromises occur.
If a programmer/designer knows they have a deadline, they usually work to that deadline, and as a result, will break that deadline.
Of course, the alternative, no deadlines, may stop the project ever finishing. A happy medium is needed.
I'm contracting at a very large computer company at the moment.. my original manager was great, and set targets, but then understood when we broke them. It worked well.
Then a new manager took over (one who needed a really good beating with the clue-stick), and started acting like a twat. The project then became late. And later. And later.
That all sounds great, but is it the first thing you see on your desktop with a message underneath saying "Connect to the Internet" (or something equally dumb but obvious)?
The Windows 98 one will give you a pretty selection of ISPs with their pros, cons and prices.
Sure, most of *us* hate it, but a dummy will love it.
How easy is installing a Linux-based system which is fully-usable. Hmmm.. I can't think of a decent way to say that.
(Before you all flame me, I don't really like Windows, but let's look at this objectively)
Once Win98 is installed, _if all goes well_, you'll have a nice little icon that helps you pick and configure an ISP. It'll pop up a nice little box when you plug your USB wossname in. It'll start the CD-ROM when you put it in the drive.
All things that we all probably hate, but all things that Joe Q. Public will _need_ if he's to use his computer more than looking what's on a disc, etc.
How easy the core operating system is to install is frankly irrelevant. How easy it is to get a useful, working, comfortable operating environment *with applications* is another matter.
Windows 98 *can* be easier than Linux in this respect (when it works!)
Re:Metered telephone calls suck so much
on
ISP War in the UK
·
· Score: 1
I was at a Q&A with some BT people about a year ago. One of the questions asked from the floor was why phone calls are metered.
Their answer (in the context of ISP services) was that if it wasn't then the phone system would get clogged with loads of modem calls, so we should think ourselves as lucky.
Like it or not, advertising exists. It also works. Much of what you see on the web (good and bad) wouldn't exist without it. Part of this is the creation of advert-based sites. These are meant to look good and feel good. You can't do that with boring old text, as in the original HTML, which was really designed for sharing of research material.
Microsoft and Netscape added those "evil" tags on demand. Companies wanted a medium to produce their pretty sites in. Nothing else existed, so they used the web. Of course, the "better" course of action would've been to design a medium for this purpose... probably using vector-based graphics and exact specification of display. But, at the moment, HTML is a layout language -- just because Tim Berners-Lee didn't intend or expect it to be used as such doesn't mean it's illegal to use it that way.
UNIX wasn't designed as a gaming platform, or to be used as a home system.. who cares?!? What can it be used for?
Many advertisers don't care about the tiny demographic of blind users, or users without the latest browser. The percentage is insignificant to the advertiser. These advertisers choose to make decisions like that, based on projected return. Many of us make our living building these sites. However, although we tend to prefer building good quality sites, sometimes we can't. The budget doesn't allow. This makes us restrict our sites to the latest browsers, or one particular branch of browsers (like Netscape).
Now what can we do about it? Fight a war between those who want the twiddles and those who don't? OR, how about building a new medium that allows both groups to co-exist? The idea of translating a bad format for compatibility reeks. GIGO.
We've got all of this noise about XML and XSL, but how much of this is designed by "webdesigners" for "webdesigners"? The concept of "building the right product, rather than building the product right" seems to be completely lost on the W3C. Until they realise that their output isn't necessarily used they way they intended, and then sit down and create something generic, flexible and powerful, we're screwed, and we're going to have this gibbering tower of babel in perpetuity.
In my opinion, the idea of separating content from presentation is a sound one. However, what won't work is disregarding the importance of presentation over content. Sometimes presentation is more important than content: e.g. for persuading PHBs. It's a sad, odious fact that PHBs control the $$$s. The $$$s pay our bills.
It's all really about target audiences... the advertisers have target audiences (and I'm afraid geeks who don't want to look at pretty sites are the minority at the moment), and the technologies have target audiences. Things like Flash have particular client target audiences, and those client target audiences have their own consumer target audiences. Use the right tool for the right job. Unfortunately, HTML is the wrong tool for any job right now. What's even worse is that there's no alternative.
(Apologies for the rambling... I'm sure this post has some valuable points hidden inside!)
I'm writing a database and a functional language, and I just bought a Powerbook G3. So there. I'd never owned (or even used extensively) a Mac before, but I'm pleasantly surprised how far they've come on recently.
It's a pretty damn good machine at a reasonable price. It's now got an X server and GNU Emacs, so I'm happy. Once OS X comes out, we'll see a few more non-Mac users moving over.
It is possible to do this with Linux/*BSD apps, though more often than not (in my experience) it involves installing from source (and taking config files to bits). When it's a lot quicker to install from an RPM or tarball, I have a tendency to think that perhaps it's not all that essential - as a result/usr/local can become a veritable maze:(
I do this quite a lot.. for software such as apache and mysql (which change reasonably often, and are worth upgrading regularly)
It's pretty useful to install different versions in different directories (such as/software/apache-1.3.4) and then symlink to the latest version (eg./software/apache)
This does need to be done from source, but it's pretty useful. A custom install is ideal for this.
I have nothing against patents per se., and even software patents in theory (things like MPEG-1 Layer 3 wouldn't have been developed without them), BUT:
a) Companies like UNISYS seem to be using them nefariously.. having the patents and then springing them on everyone eight years later.
b) They're meant to be for real innovation.. how long before the "do..while" loop gets patented? Too many obvious ideas are being patented, with existing prior art.
c) It's too prohibitively expensive for the average joe to get AND defend a patent.
There wouldn't be so many software patent applications if it wasn't so damn easy to get one. If it was clear that wide-ranging patents would be denied, it wouldn't be so common. THEN, the patent examiners could do their job properly, and examine the ones that should get patents -- ones that took many man-years to develop the idea.
I mean, that patent on bookmarks, or the thing on stylesheets. Both bloody obvious. Yes, I can say that with hindsight, since enough other people managed to come up with it independent of the patent.
The big companies (oh, like NCR and Netscape) are using them as legal ammunition... that's all.
The way patents are meant to be awarded is that the patent examiner examines the patent to see whether the "inventor" has the right to receive the patent -- that they did actually invent it.
Right now, the examiners are just rubber-stamping the patents, and then letting courts decide.
This transfers the burden of demonstrating the uniqueness of the material to the defendent. This is a really bad move.
Not all of us software authors (and other "inventors") have the capability to pay for the legal support while proving that the code we wrote should be rightfully free for us to use.
I'm thinking that the motto of the USPTO should be "Fuck the little guy."
I'm afraid to say it, but I've been thinking this would be a really good idea for ages.
One thing that Microsoft Office is in dire need of is a *decent* integrated diagram mechanism. Visio is a great product, but it needs a better interface. Put them together, and it might actually do some good.
Let's just hope they don't assimilate it and just "improve" the Office drawing tools and squash Visio in the process.
I got my big box of Lego out of my parents' attic a year or so ago.. great fun.
It *was* so much better in the old days.. you only had a few colors, and a few generic shapes, and you had to be inventive. Now, they just mould the toy in two big parts and construction consists of pushing the two parts together! That's not Proper Lego.
I'm a bit out-of-date on this, but using a server accelerator like Squid can help in a big way. Since Apache processes can be a bit big with things like mod_perl and mod_ssl running, it's best to delegate the long-distance network transfer phase of the hit to another program.
In other words, you only use apache for local hits taken by a local proxy. The proxy then transfers the data to the client. This way, the amount of time apache spends working on a hit is lowered, giving more clients/sec to apache.
This is sometimes true. I'm quite happy getting paid a tonne more than I was as a permie.
However, it's never so cut-and-dry.
I was working (permanent) for a small firm six months ago. I'd been there for three months. After a ghastly cock-up (or ten) on the part of the directors which I tried to prevent, I was blamed and given the boot. (Okay, let's deploy a brand new SMP machine as a live production box to an unattended site 60 miles away, when we've only done one day of testing, we've never had an SMP machine, and we're just about to experience a whopping great slashdot effect. I say, "Bad idea", it crashes, they give me the boot )
It was meant to be two months' severance pay, but I didn't see that either. Fired as a day's notice.
Now I'm a contractor, I feel I have *better* job security! It's less stressful, and it's a much better working environment. And I get paid more. And I'm my own boss.
It depends completely on what the working environment is like. A good contractor can work better and be more motivated than an employee.
Man, this could get more complicated than the U.S.'s electoral college system (see Sept 6th entry at www.memepool.com).
Does anyone else think that this whole thing is getting a bit too complex?
When it comes down to it, Slashdot hasn't fundamentally changed that much since it became big 'n' popular. Sorry to get back to software engineering, but wouldn't it be a good plan to sit down and do a whole new requirements analysis and design for Slashdot, based on how it is used and abused right now?
Rob's done a great job, but I think it's starting to look like a complete concept design might be a good plan. Then we might find that this moderation and meta-moderation system is unnecessary.
(I'm sure this has been said before, so please feel free to moderate me down! =) )
Well, I'm about to buy an iBook for something like this. I want something a nice easy terminal for browsing while watching TV sitting on my nice comfy sofa, rather than having to sit down at my PC.
I'd rather not have to sit upright at my PC for hours on end reading documentation... having a pad like this would make that task a lot easier.
Slashdot Mirror
To top it off, it really proves the point to see that Little Billy is drooling in the interview about one of the 'up and coming' technologies He is working on - Speech Recognition!
To give him credit, in the interview, he said something to the effect of "the up and coming technologies we're involved with" in reference to speech recognition.
His complete failure to answer any questions in that interview pissed me off, though. Every single question was answered with an advert for Microsoft, whether it was related to Microsoft or not.
And Paxman didn't badger him! I expected him to nail Billy-boy down to an answer!
Sometimes Paxman does wimp out.. he's better at scaring politicians. We'll see what he does on Sunday then.
There was a good Jon Snow interview with Billy-Boy a few months back. Snow did a pretty nice job of making Bill out to be the devil. Let's see how well Paxman does it.
Yeah, it was something to the effect of:
"Ten hours of peak-time usage of Freeserve will cost you 24 ukp, whereas it'll only cost 16 ukp on AOL"
It just sounded desperate to me. One more reason to turn AOL CD-ROMs into large-scale Tazos.
I just have a big telly instead.. even with the black bars, it's still bigger than most widescreen TVs, and is a lot cheaper (although okay, it's lower total resolution) With digital TV and DVD we get the choice of pan-n-scan or letterbox, so I don't think it really matters. Until the widescreen sets are at price parity with 4:3, it won't really take off.
Of course, we don't necessarily want a TV with the clarity of a VGA monitor... the fuzziness hides the pixelisation, especially with those crappy low-bandwidth MPEG streams from digital terrestrial.
I think that's the key to it... as soon as there's a deadline, compromises occur.
If a programmer/designer knows they have a deadline, they usually work to that deadline, and as a result, will break that deadline.
Of course, the alternative, no deadlines, may stop the project ever finishing. A happy medium is needed.
I'm contracting at a very large computer company at the moment.. my original manager was great, and set targets, but then understood when we broke them. It worked well.
Then a new manager took over (one who needed a really good beating with the clue-stick), and started acting like a twat. The project then became late. And later. And later.
That all sounds great, but is it the first thing you see on your desktop with a message underneath saying "Connect to the Internet" (or something equally dumb but obvious)?
The Windows 98 one will give you a pretty selection of ISPs with their pros, cons and prices.
Sure, most of *us* hate it, but a dummy will love it.
Well, sure.. installing Linux isn't hard.
How easy is installing a Linux-based system which is fully-usable. Hmmm.. I can't think of a decent way to say that.
(Before you all flame me, I don't really like Windows, but let's look at this objectively)
Once Win98 is installed, _if all goes well_, you'll have a nice little icon that helps you pick and configure an ISP. It'll pop up a nice little box when you plug your USB wossname in. It'll start the CD-ROM when you put it in the drive.
All things that we all probably hate, but all things that Joe Q. Public will _need_ if he's to use his computer more than looking what's on a disc, etc.
How easy the core operating system is to install is frankly irrelevant. How easy it is to get a useful, working, comfortable operating environment *with applications* is another matter.
Windows 98 *can* be easier than Linux in this respect (when it works!)
I was at a Q&A with some BT people about a year ago. One of the questions asked from the floor was why phone calls are metered.
Their answer (in the context of ISP services) was that if it wasn't then the phone system would get clogged with loads of modem calls, so we should think ourselves as lucky.
Go figure.
Like it or not, advertising exists. It also works. Much of what you see on the web (good and bad) wouldn't exist without it. Part of this is the creation of advert-based sites. These are meant to look good and feel good. You can't do that with boring old text, as in the original HTML, which was really designed for sharing of research material.
Microsoft and Netscape added those "evil" tags on demand. Companies wanted a medium to produce their pretty sites in. Nothing else existed, so they used the web. Of course, the "better" course of action would've been to design a medium for this purpose... probably using vector-based graphics and exact specification of display. But, at the moment, HTML is a layout language -- just because Tim Berners-Lee didn't intend or expect it to be used as such doesn't mean it's illegal to use it that way.
UNIX wasn't designed as a gaming platform, or to be used as a home system.. who cares?!? What can it be used for?
Many advertisers don't care about the tiny demographic of blind users, or users without the latest browser. The percentage is insignificant to the advertiser. These advertisers choose to make decisions like that, based on projected return. Many of us make our living building these sites. However, although we tend to prefer building good quality sites, sometimes we can't. The budget doesn't allow. This makes us restrict our sites to the latest browsers, or one particular branch of browsers (like Netscape).
Now what can we do about it? Fight a war between those who want the twiddles and those who don't? OR, how about building a new medium that allows both groups to co-exist? The idea of translating a bad format for compatibility reeks. GIGO.
We've got all of this noise about XML and XSL, but how much of this is designed by "webdesigners" for "webdesigners"? The concept of "building the right product, rather than building the product right" seems to be completely lost on the W3C. Until they realise that their output isn't necessarily used they way they intended, and then sit down and create something generic, flexible and powerful, we're screwed, and we're going to have this gibbering tower of babel in perpetuity.
In my opinion, the idea of separating content from presentation is a sound one. However, what won't work is disregarding the importance of presentation over content. Sometimes presentation is more important than content: e.g. for persuading PHBs. It's a sad, odious fact that PHBs control the $$$s. The $$$s pay our bills.
It's all really about target audiences... the advertisers have target audiences (and I'm afraid geeks who don't want to look at pretty sites are the minority at the moment), and the technologies have target audiences. Things like Flash have particular client target audiences, and those client target audiences have their own consumer target audiences. Use the right tool for the right job. Unfortunately, HTML is the wrong tool for any job right now. What's even worse is that there's no alternative.
(Apologies for the rambling... I'm sure this post has some valuable points hidden inside!)
I'm writing a database and a functional language, and I just bought a Powerbook G3. So there. I'd never owned (or even used extensively) a Mac before, but I'm pleasantly surprised how far they've come on recently.
It's a pretty damn good machine at a reasonable price. It's now got an X server and GNU Emacs, so I'm happy. Once OS X comes out, we'll see a few more non-Mac users moving over.
It is possible to do this with Linux/*BSD apps, though more often than not (in my experience) it involves installing from source (and taking config files to bits). When it's a lot quicker to install from an RPM or tarball, I have a tendency to think that perhaps it's not all that essential - as a result /usr/local can become a veritable maze :(
/software/apache-1.3.4) and then symlink to the latest version (eg. /software/apache)
I do this quite a lot.. for software such as apache and mysql (which change reasonably often, and are worth upgrading regularly)
It's pretty useful to install different versions in different directories (such as
This does need to be done from source, but it's pretty useful. A custom install is ideal for this.
I have nothing against patents per se., and even software patents in theory (things like MPEG-1 Layer 3 wouldn't have been developed without them), BUT:
a) Companies like UNISYS seem to be using them nefariously.. having the patents and then springing them on everyone eight years later.
b) They're meant to be for real innovation.. how long before the "do..while" loop gets patented? Too many obvious ideas are being patented, with existing prior art.
c) It's too prohibitively expensive for the average joe to get AND defend a patent.
There wouldn't be so many software patent applications if it wasn't so damn easy to get one. If it was clear that wide-ranging patents would be denied, it wouldn't be so common. THEN, the patent examiners could do their job properly, and examine the ones that should get patents -- ones that took many man-years to develop the idea.
I mean, that patent on bookmarks, or the thing on stylesheets. Both bloody obvious. Yes, I can say that with hindsight, since enough other people managed to come up with it independent of the patent.
The big companies (oh, like NCR and Netscape) are using them as legal ammunition... that's all.
Agreed.. the current problem is the USPTO.
The way patents are meant to be awarded is that the patent examiner examines the patent to see whether the "inventor" has the right to receive the patent -- that they did actually invent it.
Right now, the examiners are just rubber-stamping the patents, and then letting courts decide.
This transfers the burden of demonstrating the uniqueness of the material to the defendent. This is a really bad move.
Not all of us software authors (and other "inventors") have the capability to pay for the legal support while proving that the code we wrote should be rightfully free for us to use.
I'm thinking that the motto of the USPTO should be "Fuck the little guy."
..or NTEmacs (now part of the Emacs dist?)
(Let's not start another Emacs -v- Vi(m) war...)
Also, this has relaxed the rules enough for big business to sell their stuff *reasonably* easy.
Thus, there's a lot less incentive for big business to continue lobbying for total freedom.
Now that big business is happy, what's the chance of this trend continuing any further?
I'm afraid to say it, but I've been thinking this would be a really good idea for ages.
One thing that Microsoft Office is in dire need of is a *decent* integrated diagram mechanism. Visio is a great product, but it needs a better interface. Put them together, and it might actually do some good.
Let's just hope they don't assimilate it and just "improve" the Office drawing tools and squash Visio in the process.
I got my big box of Lego out of my parents' attic a year or so ago.. great fun.
It *was* so much better in the old days.. you only had a few colors, and a few generic shapes, and you had to be inventive. Now, they just mould the toy in two big parts and construction consists of pushing the two parts together! That's not Proper Lego.
I'm a bit out-of-date on this, but using a server accelerator like Squid can help in a big way. Since Apache processes can be a bit big with things like mod_perl and mod_ssl running, it's best to delegate the long-distance network transfer phase of the hit to another program.
In other words, you only use apache for local hits taken by a local proxy. The proxy then transfers the data to the client. This way, the amount of time apache spends working on a hit is lowered, giving more clients/sec to apache.
Anyone else know more about this?
I should bloody well hope it works. It's my birthday. =)
This is sometimes true. I'm quite happy getting paid a tonne more than I was as a permie.
However, it's never so cut-and-dry.
I was working (permanent) for a small firm six months ago. I'd been there for three months. After a ghastly cock-up (or ten) on the part of the directors which I tried to prevent, I was blamed and given the boot. (Okay, let's deploy a brand new SMP machine as a live production box to an unattended site 60 miles away, when we've only done one day of testing, we've never had an SMP machine, and we're just about to experience a whopping great slashdot effect. I say, "Bad idea", it crashes, they give me the boot )
It was meant to be two months' severance pay, but I didn't see that either. Fired as a day's notice.
Now I'm a contractor, I feel I have *better* job security! It's less stressful, and it's a much better working environment. And I get paid more. And I'm my own boss.
It depends completely on what the working environment is like. A good contractor can work better and be more motivated than an employee.
Man, this could get more complicated than the U.S.'s electoral college system (see Sept 6th entry at www.memepool.com).
Does anyone else think that this whole thing is getting a bit too complex?
When it comes down to it, Slashdot hasn't fundamentally changed that much since it became big 'n' popular. Sorry to get back to software engineering, but wouldn't it be a good plan to sit down and do a whole new requirements analysis and design for Slashdot, based on how it is used and abused right now?
Rob's done a great job, but I think it's starting to look like a complete concept design might be a good plan. Then we might find that this moderation and meta-moderation system is unnecessary.
(I'm sure this has been said before, so please feel free to moderate me down! =) )
Well, I'm about to buy an iBook for something like this. I want something a nice easy terminal for browsing while watching TV sitting on my nice comfy sofa, rather than having to sit down at my PC.
I'd rather not have to sit upright at my PC for hours on end reading documentation... having a pad like this would make that task a lot easier.
Found an old version in my mailbox. Will replace it with the newer version if/when I find it:
http://www.litebase.com/witm.el
Comments/enhancements welcomed.
Somewhere round here I've got "matrix-mode" for Emacs. Couldn't quite get the kanji characters working though.
It's on my desktop somewhere.. =)
I don't know about anyone else, but I think "Microsoft" hardware tends to be pretty good (since it's all subcontracted out)
Okay, apart from that sodding Barney Actimate thing. I'm thinking about buying one of those just so I can set fire to it.