One of our production systems averages around 500/s, with peaks at over 3000/s, with roughly that sort of table structure... it's a dual Opteron, H/W RAID10 across two channels, 6 disks and 4GB of RAM and a separate SATA WAL drive. Thing is, it's not just an insert, it's a call to a stored procedure that does some data hygiene, then a select and then either an insert or update depending on the result of the select. Very fast, very happy!
Although the source for the reference platform has been available for some time, the fact that it may become 'free' means forks are inevitable, and that's the only thing that's missing from Java, namely the freedom to fork it. Mind you, if the C++ crowd get hold of it that's what it will be... completely forked.
If I recall correctly, 80 MIPS is more than a 486 can turn out - and I'm certain abstraction layers have been used on far lesser CPU's. As for RISC, I still maintain an AlphaStation 4/166 from 1993 that does, well, 166 MIPS - but it runs RH7.2 very nicely, is a web proxy for 25 people on a 1Mb pipe, supports at least two simultaneous X clients on some days and still has time left over to run the odd web log analysis. So, if 80 MIPS is not enough for you, I would hate to be the the poor sod who has to live with your software!
I have just realised that I made a mistake. The report mentioned was not that from Microsoft, but from the council's chief executive. I retract my previous comment. I still think the whole thing stinks, though.
Decision:
The Mayor in consultation with the Cabinet agreed:
...
(iv) to agree to waive Standing Order 8 (requirements for all contracts over £25,000 to be subject to formal tendering) to enable the Strategic Partnership Agreement with Microsoft to be concluded despite the absence of any formal tendering for the reasons set out in the report;
...
(viii) to note the information in the exempt report related to this item.
So, it seems that MS bypassed the usual tendering process by means of their own funded report... and the council can't make public the findings of their experience.
There are many, in fact. Most are embedded in obscure products. Often they involve navigational access. Check out Equinox from Compsoft plc and you will see what I mean.
I really hope Novell do well on Linux, it is the kind of OS that they should have had all along which would have let them concentrate on what they do best.
I recently wrote a series of reports for my boss in which I had to make the cases for and against a port of our product (an app server) to Netware. Bear in mind that it was over two years ago now that we decided to discontinue Netware development because sales had dropped through the floor (it was, and still is, available in older versions which we are quite happy to sell and support). Apparently, our sales department discovered that there was still some residual demand which was affecting their commission. The idea was shot down by the board after only a few seconds' reflection - apparently, my boss (Technical Director) didn't even have to present our work on the matter, they just knew it was a daft idea!
Is anyone actually considering picking up Netware development for commercial products? I bet there is not a single one out there. Prove me wrong.
Netware has become a legacy faster than anything I have ever seen - even the mainframe concept will outlive it. Netware is dead, long live Novell.
"I don't see any fundamental reason why science should be unable to create something more capable than the products of evolution if given enough time."
There isn't enough time - nor has mankind the wit.
The stuff of our brains is not digital, serial or silicon - it is chemical, analogue, noisy and random. We will never emulate ourselves with our inventions.
It strikes me that the complaint that it just turns people into robots only highlights that the people that are replaced by robots are actually only doing jobs that can be replaced by robots... I think this means that those jobs should defininately be done with robots wherever possible to avoid the boring, repetitive strain on people (ie: we shouldn't allow people to do jobs that robots can do because people aren't robots and shouldn't be treated as such). I think I made sense there... not sure!
That thought came about because this system can only be applied in circumstances where a robot is also suitable - quite frankly, if a vison and and help system knows what to do - it can drive a robot.
Your points are valid, but I don't understand you. The vast majority of businesses are very much smaller than Ford (certainly less than 50 people) and this means they actually have to be nice to their customers and employees as the relative levels of investment and exposue are so much more critical. Forget the big companies, they will always be eating their young in order to 'win' - the rest of the world has young to feed.
"Itâ(TM)s no secret. To win at business, you must perform better than your competition. Better. Stronger. Faster." -
Sometimes, you americans really lose sight of the facts.
Firstly, there is no 'winner' in business; at any one time, in any given market, on some particular scale, someone is ahead of their rivals - but it never stays that way for very long.
Secondly, is 'winning' actually the point of being in business? Surely it must be just as important to look after your staff and customers? Without them, you can't 'win'.
To use a pair of americanisms, give us a break from this total bullshit.
...European Patent Convention? Article 52 is particularly interesting. Americans, take note: European patent law has been harmonised for nearly 30 years and specifically bans software patents as you know them.
Don't be silly. Just go to one European country at a time. When you are comfortable, go to another. Start with one that uses the same language as you, perhaps Britain or France... now, that didn't hurt, did it? Can't be any harder than cracking Japan, eh?
Re:Stop the anti-MS BS all the damned time
on
Hijacking .NET
·
· Score: 1
Actually, it is a security issue, but only because the vast majority of programmers will unwittingly treat private members as secure. Programming languages, like spoken languages, are hostage to the skills of their users. MS know this, but they choose to ignore the fact that nearly all programmers will fail to have a deep understanding of these features and gave them the tools to make weak programs. Let's consider pointers in C# for a moment... hmm, shite strikes twice in the same place, methinks.
Slashdot Mirror
I'm terribly sorry, but I really do find it amusing when Amercians whine that the world isn't revolving around them.
Nice post, Rogerborg.
Sam, I'm sure you'll remember Richard's laptop starting with 'Good Morning Vietnaaaaaam' every day... need I say more?
Matt.
Yup... that's told 'em.
The good doctor breaks a longevity record every second he lives... I'll have whatever he's having!
Matt.
One of our production systems averages around 500/s, with peaks at over 3000/s, with roughly that sort of table structure... it's a dual Opteron, H/W RAID10 across two channels, 6 disks and 4GB of RAM and a separate SATA WAL drive. Thing is, it's not just an insert, it's a call to a stored procedure that does some data hygiene, then a select and then either an insert or update depending on the result of the select. Very fast, very happy!
Although the source for the reference platform has been available for some time, the fact that it may become 'free' means forks are inevitable, and that's the only thing that's missing from Java, namely the freedom to fork it. Mind you, if the C++ crowd get hold of it that's what it will be... completely forked.
...then we'll see who should write software.
Just 'cos I beat you to the sentiment Nah Nah Ne Nah Nah! (oh, it's still your turn to offer something constructive or supportive, btw).
I will lend my lower user ID to you by endorsing every word in your post. If I had mod points today, I would lend them too.
Dogs don't kill people, Wabbits do!
If I recall correctly, 80 MIPS is more than a 486 can turn out - and I'm certain abstraction layers have been used on far lesser CPU's. As for RISC, I still maintain an AlphaStation 4/166 from 1993 that does, well, 166 MIPS - but it runs RH7.2 very nicely, is a web proxy for 25 people on a 1Mb pipe, supports at least two simultaneous X clients on some days and still has time left over to run the odd web log analysis. So, if 80 MIPS is not enough for you, I would hate to be the the poor sod who has to live with your software!
I have just realised that I made a mistake. The report mentioned was not that from Microsoft, but from the council's chief executive. I retract my previous comment. I still think the whole thing stinks, though.
There are many, in fact. Most are embedded in obscure products. Often they involve navigational access. Check out Equinox from Compsoft plc and you will see what I mean.
"hope it actually works before somebody is on trial for his life though..."
Which is why capital punishment should be abandoned - and not just for this particular example.
I really hope Novell do well on Linux, it is the kind of OS that they should have had all along which would have let them concentrate on what they do best.
I recently wrote a series of reports for my boss in which I had to make the cases for and against a port of our product (an app server) to Netware. Bear in mind that it was over two years ago now that we decided to discontinue Netware development because sales had dropped through the floor (it was, and still is, available in older versions which we are quite happy to sell and support). Apparently, our sales department discovered that there was still some residual demand which was affecting their commission. The idea was shot down by the board after only a few seconds' reflection - apparently, my boss (Technical Director) didn't even have to present our work on the matter, they just knew it was a daft idea!
Is anyone actually considering picking up Netware development for commercial products? I bet there is not a single one out there. Prove me wrong.
Netware has become a legacy faster than anything I have ever seen - even the mainframe concept will outlive it. Netware is dead, long live Novell.
"I don't see any fundamental reason why science should be unable to create something more capable than the products of evolution if given enough time."
There isn't enough time - nor has mankind the wit.
The stuff of our brains is not digital, serial or silicon - it is chemical, analogue, noisy and random. We will never emulate ourselves with our inventions.
It strikes me that the complaint that it just turns people into robots only highlights that the people that are replaced by robots are actually only doing jobs that can be replaced by robots... I think this means that those jobs should defininately be done with robots wherever possible to avoid the boring, repetitive strain on people (ie: we shouldn't allow people to do jobs that robots can do because people aren't robots and shouldn't be treated as such). I think I made sense there... not sure!
That thought came about because this system can only be applied in circumstances where a robot is also suitable - quite frankly, if a vison and and help system knows what to do - it can drive a robot.
...it'll leak in your pocket, making it just as messy as a pen.
Your points are valid, but I don't understand you. The vast majority of businesses are very much smaller than Ford (certainly less than 50 people) and this means they actually have to be nice to their customers and employees as the relative levels of investment and exposue are so much more critical. Forget the big companies, they will always be eating their young in order to 'win' - the rest of the world has young to feed.
"Itâ(TM)s no secret. To win at business, you must perform better than your competition. Better. Stronger. Faster." -
Sometimes, you americans really lose sight of the facts.
Firstly, there is no 'winner' in business; at any one time, in any given market, on some particular scale, someone is ahead of their rivals - but it never stays that way for very long.
Secondly, is 'winning' actually the point of being in business? Surely it must be just as important to look after your staff and customers? Without them, you can't 'win'.
To use a pair of americanisms, give us a break from this total bullshit.
...European Patent Convention? Article 52 is particularly interesting. Americans, take note: European patent law has been harmonised for nearly 30 years and specifically bans software patents as you know them.
Don't be silly. Just go to one European country at a time. When you are comfortable, go to another. Start with one that uses the same language as you, perhaps Britain or France... now, that didn't hurt, did it? Can't be any harder than cracking Japan, eh?
Actually, it is a security issue, but only because the vast majority of programmers will unwittingly treat private members as secure. Programming languages, like spoken languages, are hostage to the skills of their users. MS know this, but they choose to ignore the fact that nearly all programmers will fail to have a deep understanding of these features and gave them the tools to make weak programs. Let's consider pointers in C# for a moment... hmm, shite strikes twice in the same place, methinks.