Well, certainly if the original website was over-engineered with mysql then it follows that adding triggers to the database takes the over-engineering to the extreme.
Its hard to tell if the original 2 month website was over-engineered or not without knowing what the customer actually asked for.
Are you confused about the difference between "quality" and "features"?
"Quality" and "features" are not exclusive.
Negative extremes of these two are "over-engineered" and "bloated"
Would you code in triggers even if your project didn't need them, or merely insist your DB had them in case you might need them? (Smells like over-engineered)
Looks like they use InterMail pop3 server (telnet pop.east.cox.net 110) and smtp server (telnet smtp.east.cox.net 25): 220 lakermmtao11.cox.net ESMTP server (InterMail vM.6.01.03.02 201-2131-111-104- 20040324) ready Mon, 24 May 2004 19:00:55 -0400 (was 4xx too busy a minute ago)
Intermail is/was produced/sold by Openwave Intermail is no longer available and support has been discontinued. For Openwave email products please visit our Email Mx page.
So, no support. Indications are that it runs on windows servers.
You can program em both in FORTH if you want to (see forth) - forth is like a macro-macro-assembler where you derive your own language in it, one which is ideally suited to your problem.
It has a camera to take photos at a moments notice (better than no photo at all) and even short movie clips with sound.
I user opera for web browsing with the nice PDA layout so I dont have to keep scrolling horizontally, and I use the email client for those times when I need mobile email.
Then there is the standard PDA calendar and phone book.
If the P900 were any smaller, the screen would me smaller, and I wouldn't want that!
The only thing that hurts are GPRS data prices at the moment, so I have to really justify "online" use.
There are always more smartphones coming out.
I preder the larger screen touch-screen mainly-PDA style smartphone, but there are plenty of more robust-screen mainly-phone style smartphones too.
I think Orange are the leaders in releasing smartphones, and I don't think thats my bias.
The problem with the recounts wasn't the counting, it was the quality of the voting MACHINES (in the UK we use a pencil) and how these problems were raised by HUMANS.
Remember recent machine counts of written polls report on slashdot failed due to bad calibration of the machine and caught only by a lucky selection for quality control tests?
Multiple people who value their freedom will generally be able to manage and detect sound elections than machines will.
We all know its possible to design secure and tamper evident voting machines -- its probably not even that hard.
What rubbish you speak. Election counting is and always has been simple.
When you get a complex system like a computer you need to be sure thats all its doing and thats all its ever doing.
When steel ballot boxes are being stored they can be stored in a warehouse. Its hard to tamper with ignorant steel boxes in a meaningful way.
To subvert thousands of humans who count ballots manually leaves, lets say, thousands of human witnesses.
When electronic voting machines are being stored they need to be watched carefully to make sure they aren't modified, don't have their guts swapped out, etc, this between-election security is also very expensive.
Its expensive before you start, its expensive to run, and expensive to store with many possible points of subversion.
It will do humans good to count votes and realise they don't want to delegate safeguarding their democracy to fickle machines.
With Aegis the baseline code "always" works; it has to pass all the build tests to become baseline.
You can't add a new feature without first defining a test for it to pass, you can't fix a bug without defining a test that the old baseline failed and the new baseline passes.
So marketing can walk up and say "release now" or "add these features" and you can do either. But you can't "release these features now" because the system won't let you.
When marketing say "release now" they can only have the bits that work. And when they say "add these features" they can only get those features when they work.
* Do what I did and instant-messenger spam all your friends and get them to donate * Skip a nights partying and donate that * Sell some stuff on ebay and donate that * Sell your friends stuff on ebay and donate that * Have a pizza party for your friends ($5 a person) but then use cheap pizza (har har!) * Drink cheap bear (ugh) and therefore less of it and donate savings to EFF * Instead of buying new clothes get a magic-marker and write on your old ones:
I could have bought new clothes but I'm paying the EFF ton right {Amazon|bad patents|etc} * hold a live auction of junk! (How novel!) That could be fun. Don't forget to sell refreshments
On a windows system I have done some development on recently (this year) there were cases where code modules needed to obtain summary information from other user systems.
The public way was to query each user system for the data, and to poll for updates.
The official secret way we had to fight to get access to consisted of being able to register for callbacks for changing values in a central data store shared by all the user systems.
That makes simpler, faster and more responsive software and increases battery life.
There are important support and "API stability" reasons why some APIs remain secret (or restricted) and MS sometimes have to weigh up the advantages and disadvantages of revealing some APIs still in development.
I don't always like their decisions but I understand the reasons behind them.
So yes, there are hidden APIs that can bring great benefit. And yes it can make code more brittle and we needed more per-target compilation configurations, but it was worth it for the speed, simplicity and battery savings in our case.
We havent managed it with spam filters and I doubt webwasher is going to solve this problem just to suit your (left aside) off-work browsing at work.
To fit your bosses needs, web-washer draws a very blunt line. It cuts out the dirt and the providers of the "dirt". You think its a special class of "dirt" that just looks like the regular dirt but thats OK really because its only enjoyed by people who have the patience to read the full story maybe? I think your boss thinks things are just fine.
When I ran an ISP and we got a demanding email to remove "all infringing links" I replied pointing out that I wouldn't be able to tell which material was posted with permission and which would not, but that they ought to be able to, and if they could tell me I would look at doing it.
The Debian installer installs the same kernel that it boots from...
Well then it missed the loadable network modules that had been loaded during installation. This is something redhat gets right with the init ram disk stools.
I think I'm a pretty good developer and very computer literate.
Jigdo isn't worth the time it takes to say it.
And why (0h why) do the debian installers insist on using stripped down kernels WITHOUT any compiled modules either? Whats the use of that?
For installation ISOs include nearly EVERY kernel module.
I had no end of trouble with debian and SID/Sarge installer BECAUSE the installation-time kernel had the right modules but the installed kernel didn't! I had to keep "recovering" from the installation CD to get a box with PCMCIA net drivers and spent ages trying to get a.deb (not udeb - what are udeb anyway) that had the right modules.
Yeah I could have downloaded the whole kernel source and compiled it on a P75 laptop with 80MB RAM but I didn't think it worthwhile.
Slashdot Mirror
Yes, but now the DARPA have finished developing it each country is paying their own share.
Thankyou DARPA for the funding and early development, very essential.
But now DARPA contribute no more to the internet than Marconi or Farnsworth do to programme broadcasting.
Unlike the BBC which develops programmes using UK license-payers money to then market abroad.
Sam
Well, certainly if the original website was over-engineered with mysql then it follows that adding triggers to the database takes the over-engineering to the extreme.
Its hard to tell if the original 2 month website was over-engineered or not without knowing what the customer actually asked for.
Sam
Are you confused about the difference between "quality" and "features"?
"Quality" and "features" are not exclusive.
Negative extremes of these two are "over-engineered" and "bloated"
Would you code in triggers even if your project didn't need them, or merely insist your DB had them in case you might need them? (Smells like over-engineered)
Sam
pop3 problems it seems with no time to repair.
Looks like they use InterMail pop3 server (telnet pop.east.cox.net 110) and
smtp server
(telnet smtp.east.cox.net 25): 220 lakermmtao11.cox.net ESMTP server (InterMail vM.6.01.03.02 201-2131-111-104-
20040324) ready Mon, 24 May 2004 19:00:55 -0400
(was 4xx too busy a minute ago)
Intermail is/was produced/sold by Openwave
Intermail is no longer available and support has been discontinued. For Openwave email products please visit our Email Mx page.
So, no support.
Indications are that it runs on windows servers.
Draw your own conclusions
Sam
USB here:
http://www.mpeltd.demon.co.uk/usbstamp.htm
Very tiny here:
http://www.mpeltd.demon.co.uk/tiniarm.htm
You can program em both in FORTH if you want to (see forth) - forth is like a macro-macro-assembler where you derive your own language in it, one which is ideally suited to your problem.
Sam
I've used the Microsoft SPV1, E200,
and Sony Ericsson P800 and P900, currently using the P900.
(You might want to read my CV, brief and full)
I have to say smartphones are the bees knees.
After all why wouldn't you want your PDA to also be a phone with internet connectivity?
My P900 is my ebook reader and offline browser, and my portable games machine
It has a camera to take photos at a moments notice (better than no photo at all) and even short movie clips with sound.
I user opera for web browsing with the nice PDA layout so I dont have to keep scrolling horizontally, and I use the email client for those times when I need mobile email.
Then there is the standard PDA calendar and phone book.
If the P900 were any smaller, the screen would me smaller, and I wouldn't want that!
The only thing that hurts are GPRS data prices at the moment, so I have to really justify "online" use.
There are always more smartphones coming out.
I preder the larger screen touch-screen mainly-PDA style smartphone, but there are plenty of more robust-screen mainly-phone style smartphones too.
I think Orange are the leaders in releasing smartphones, and I don't think thats my bias.
Sam
The problem with the recounts wasn't the counting, it was the quality of the voting MACHINES (in the UK we use a pencil) and how these problems were raised by HUMANS.
Remember recent machine counts of written polls report on slashdot failed due to bad calibration of the machine and caught only by a lucky selection for quality control tests?
Multiple people who value their freedom will generally be able to manage and detect sound elections than machines will.
Sam
Quite so, thanks for picking up on that.
Sam
What rubbish you speak. Election counting is and always has been simple.
When you get a complex system like a computer you need to be sure thats all its doing and thats all its ever doing.
When steel ballot boxes are being stored they can be stored in a warehouse. Its hard to tamper with ignorant steel boxes in a meaningful way.
To subvert thousands of humans who count ballots manually leaves, lets say, thousands of human witnesses.
When electronic voting machines are being stored they need to be watched carefully to make sure they aren't modified, don't have their guts swapped out, etc, this between-election security is also very expensive.
Its expensive before you start, its expensive to run, and expensive to store with many possible points of subversion.
It will do humans good to count votes and realise they don't want to delegate safeguarding their democracy to fickle machines.
Sam
With Aegis the baseline code "always" works; it has to pass all the build tests to become baseline.
You can't add a new feature without first defining a test for it to pass, you can't fix a bug without defining a test that the old baseline failed and the new baseline passes.
So marketing can walk up and say "release now" or "add these features" and you can do either. But you can't "release these features now" because the system won't let you.
When marketing say "release now" they can only have the bits that work. And when they say "add these features" they can only get those features when they work.
Sam
Gardeners have been growing plants in greenhouses and flowerbeds for years; why shouldn't a businessman grow a business?
Sam
* Do what I did and instant-messenger spam all your friends and get them to donate
* Skip a nights partying and donate that
* Sell some stuff on ebay and donate that
* Sell your friends stuff on ebay and donate that
* Have a pizza party for your friends ($5 a person) but then use cheap pizza (har har!)
* Drink cheap bear (ugh) and therefore less of it and donate savings to EFF
* Instead of buying new clothes get a magic-marker and write on your old ones:
I could have bought new clothes but I'm paying the EFF ton right {Amazon|bad patents|etc}
* hold a live auction of junk! (How novel!) That could be fun. Don't forget to sell refreshments
Sam
This is a worthy cause!
Donate now, because its CHEAPER than joining a defense fund later.
https://secure.eff.org/
I just joined! Thankyou EFF for taking up this cause!
Sam
Sam
Hey,
notice I restricted my comments to the irony and didn't try to impose anyone alses religious system, or respect for othes deitys, upon you.
Your extended excuse indicates some guilt though.
Sam
On a windows system I have done some development on recently (this year) there were cases where code modules needed to obtain summary information from other user systems.
The public way was to query each user system for the data, and to poll for updates.
The official secret way we had to fight to get access to consisted of being able to register for callbacks for changing values in a central data store shared by all the user systems.
That makes simpler, faster and more responsive software and increases battery life.
There are important support and "API stability" reasons why some APIs remain secret (or restricted) and MS sometimes have to weigh up the advantages and disadvantages of revealing some APIs still in development.
I don't always like their decisions but I understand the reasons behind them.
So yes, there are hidden APIs that can bring great benefit. And yes it can make code more brittle and we needed more per-target compilation configurations, but it was worth it for the speed, simplicity and battery savings in our case.
Sam
Quite Ironic. The respectful upper-case J and C combined with the disrespectful use of the name.
If webwasher could do it so could spam filters.
We havent managed it with spam filters and I doubt webwasher is going to solve this problem just to suit your (left aside) off-work browsing at work.
To fit your bosses needs, web-washer draws a very blunt line. It cuts out the dirt and the providers of the "dirt". You think its a special class of "dirt" that just looks like the regular dirt but thats OK really because its only enjoyed by people who have the patience to read the full story maybe? I think your boss thinks things are just fine.
So just don't use webwasher at home.
Sam
When I ran an ISP and we got a demanding email to remove "all infringing links" I replied pointing out that I wouldn't be able to tell which material was posted with permission and which would not, but that they ought to be able to, and if they could tell me I would look at doing it.
Sam
I understood one reason for the mirror was than JANET (Joint Academic NETwork) were being billed per byte for UK-bound transatlantic traffic.
But reading the article (shock - its a small article) it seems like what was turned down was an enhanced mirror, personalisable, RSS based, etc.
They still enterain hopes of running the mirror service "perhaps on a smaller scale".
Sam
what does "reliable" mean?
Its the weather thats not reliable.
And the intelligence
And the double agents
And the cowards
And the enemy
And the allies
Perhaps "useful" is what is needed.
Sam
Well then it missed the loadable network modules that had been loaded during installation. This is something redhat gets right with the init ram disk stools.
Sam
I'm still going with debian for now, I just can't fathom some of the decisions.
Sam
I think I'm a pretty good developer and very computer literate.
.deb (not udeb - what are udeb anyway) that had the right modules.
Jigdo isn't worth the time it takes to say it.
And why (0h why) do the debian installers insist on using stripped down kernels WITHOUT any compiled modules either? Whats the use of that?
For installation ISOs include nearly EVERY kernel module.
I had no end of trouble with debian and SID/Sarge installer BECAUSE the installation-time kernel had the right modules but the installed kernel didn't! I had to keep "recovering" from the installation CD to get a box with PCMCIA net drivers and spent ages trying to get a
Yeah I could have downloaded the whole kernel source and compiled it on a P75 laptop with 80MB RAM but I didn't think it worthwhile.
Sam
Tesco have the book at a fair price, UKP 13.88
Price runner shows that Amazon are the cheapest but they are still ut of favour for aggressive and stupid patent practice, so pick another supplier from the list