Looking some more on the www.tulip.nl site, it turns out that it really is like I described before.
The current management have cut away everything that made Tulip a computer manufacturer, and only left a small bureau that defends some patents, trademarked brand names, and "markets".
A news article on the site shows a clear intent to go the SCO way: not design, sell or make any product, but just cash in on existing names and try to get money from others who are actually trying to do some innovation (and may unsuspectingly thread on something you have registered in the previous century).
The manager is proud of it, and enjoys the good life in a house in France, a fast car, and an office in a classic old city center.
Well, what should we say... it is a way to approach things, but I doubt if it will ever make them as successfull as e.g. Dell.
>The company has been on the brink of bankrupcy a number of times.
Probably again... They show all the characteristics: digging through their old stack of patents and finding violations, and now looking in the pile of "brand names" they own and trying to cash-in on those.
From the entire article it is apparent that they expect nothing less than a steady stream of royalty money coming in all by itself by just declaring "commodore is our brand name", fighting all people who setup sites of their own, and bringing out some software emulator for the PC that they blindly assume 6 million people will buy from them.
The pirates that abuse the Usenet system to transport their so-called "Warez" already have such a system in place. The loss of part of the data can be overcome by extra (R-S) redundancy added to it when it was sent.
The reason they are doing this, is that the price level outside the US is usually higher.
Dell and Compaq sell systems all over the world, and when one would convert the US price plus shipping and tax to local currency, it would often be less than the local price from them.
So, to prevent people taking advantage of the world wide web and order their system from the US, they put some clauses into the contract to prevent you from doing that.
Of course this is protectionism, and in many countries it is outlawed. But they get away with it, apparently.
A lot of things are illegal in Holland, but nobody ever bothers to do something about it. You may know about cannabis posession. It isn't legal but no policeman or judge will bother.
What they did here is certainly illegal. However, they will most likely get away with it. Even when the original owners of the domain speak up, they will most likely only hear that cases like this are not a priority of the legal system.
(the priorities of the legal system are very unclear. there is a definite bias towards cases that yield a good income, like putting up a camouflaged speed trap on a motorway behind a "30km because of roadworks" sign, or towing away cars from places where it not allowed to park them. finding and locking away the thief of your bicycle or laptop isn't a priority. hence, plain citizens are a victim but criminals rarely are)
>If you ask me (which nobody has), we should be using lower frequencies, not higher frequencies. Sure, lower frequencies require more bandwidth for the same speeds, but with lower bandwidths, obstacles wouldn't be a big deal, and you could transmit far, far further with less power as well.
We use Mozilla at work. I did almost all of the deployment stuff. It is *completely* automatic. When logged on, PC's get the newest version of Mozilla (both Windows 95 and Windows 2000), and it is configured for the logged-on user completely automatically. The user never has to type a thing, except their mail password.
How is this done? Well, indeed you have to overcome some of the nasty features of the configuration manager. Using some tricks it is possible to get rid of the salt, and to store the Mozilla profile inside the User's roaming profile, so that they don't have to choose the profile when Mozilla is launched. (every user gets their own setup of Mozilla, including mail address, addressbook, and all other preferences, as determined by their logon)
The first trick is to remove the profile, create a new profile named "default", change its directory to a location that is the same for each user (network drive, profile directory via a SUBST) and save that. Keep the registry.dat file that has been created in a safe place, and copy it over at every login. This resolves the salt problem and keeps your profile in the same place for every user. To setup a new user, take the prefs.js of a standard user, delete all lines that refer to the user (mail address etc) and save that as a base config. Write a simple script that takes this file, appends the user-specific lines, and creates prefs.js. Then your users don't need to config anything.
For calendar, we use Maorong Zou's "WebCalendar", a web application running on a Linux server. It is not local to the browser, but it works well.
So far, it is all very nice. There are of course problems as well:
- Users complain it is slow. Especially the group of users that still has Pentium-class 64MB systems. It sure could be faster.
- There is a performance problem in the IMAP code. Downloading large attachments over the LAN is painfully slow, even on very fast systems. This is an open bug.
- There sometimes are problems with printing. Mozilla re-loads the page when printing it, sometimes several times. It often crashes when complicated javascript is in the page to be printed.
- There is an irritating problem when opening a new window, where it tends to re-use existing windows that contain valuable info (like your calendar). Workaround is to use tabs instead of windows.
Of course I can go on for a long time on the topic of using Mozilla, the advantages, the disadvantages, the tricks of installation and configuration, etc. But overal I am quite happy. Maybe not all the users are, but that usually is a matter of "I am accustomed to Outlook and IE". Those people are usually also accustomed to viruses.
>It is a fairly trivial matter for most regular/. readers to back trace a spam mail to the source server. In nearly all cases the server is an open relay or has been owned - either way the plug should be pulled.
I think you have not looked at the matter last year. What you say may have been true in the past, but the spammer's tactics have changed. They use proxies now, not relays. There is no way to trace the path back to them, for a regular/. reader.
You would need co-operation from the access provider of an "innocent" family using cable or adsl internet, and from that family.
Well, in fact even from losers like the author of AnalogX Proxy and other Windows proxies that are by default open to the Internet and do not log.
>Is there any simple (as in Joe User simple, not simple as in run this script, patch this file, compile this kernal simple) way to get WinModem support under Linux?
How long until telephone modem support is just as low a priority as floppy support?
The floppy driver has been terrible for many years now, and nobody seems to care. Who uses floppies anyway?
This is not a difference with Windows. Most people cannot install Windows either. They get it installed on their new PC. When they would have to install it, let it find hardware, and install drivers all from scratch they would find that daunting too.
In this filter, executable files are not only recognized by their extension (and.msw would be very easily added to the list) but also by the first couple of bytes of the file. All the processor-executable files have the same header structure in Windows, and are thus very easily identified.
With every next worm, I wonder why there does not appear one that first propagates and then erases all data it can touch. You know, like the good old days when there was supposed to be a data in the near future when all PCs in the universe would crash because of a virus. (quickly purchase a virus scanner or you will be doomed)
The message seems to be coming from a friend, has an attachment that promises to be a document, when you unzip it it contains a file named like a document, so the normal next step would be to doubleclick on it, expecting it to be opened.
That this means "run it" in this case is a distinction that has been blurred by Windows.
We have no choice what we buy. We get Windows as part of the workstation (the Microsoft Tax), it would be foolish to pay the tax and then buy another version and run that...
But it is good to hear that something is being done about it. With Linux, we don't have that problem. (all the languages are on one set of CDs)
Actually it is a big shame that there are still language-dependencies in the OS, so that you need a different version of patches for each language. (same for drivers etc)
We run English versions on the servers (for legacy reasons and because some things just aren't available in all languages), and have Dutch Windows 2000 on the workstations. Joy, joy, joy...
Probably we will have to wait a while for SP4.
Re:What can we reasonably expect from bug reports?
on
Mozilla 1.4 RC3 Is Out
·
· Score: 1
>However, we do most definitely appreciate the effort put into submitting clean bug reports, and whenever possible try to follow-up on them.
I certainly do not want to criticize the tremendous work done by a lot of volunteers. When I did not have a lot of projects myself, I would probably be able to assist... but for this application, I am merely a consumer that is willing to report problems and spend some time describing a test case. Of course, it helps when knowing that actually something is going to improve because of these reports.
Re:What can we reasonably expect from bug reports?
on
Mozilla 1.4 RC3 Is Out
·
· Score: 1
The two bugs I reported myself are 207607 and 207611. Slow IMAP is reported in some different bugs, but the main one is 147285. Someone is working on that now, it seems.... but too late for 1.4 I fear.
>I really think that I should write a filter that spell-checks an email, and rejects it if over 50% of the words with 5 or more letters are misspelled.
That is a good idea. Of course it will need a large dictionary of correctly-spelled words in a couple of languages.
Maybe it is also possible to create an algorithm that decides if a string of characters is likely to be a valid word in a western language. I.e. that separates the correct words from the hwreqwuir. That could be used like the spellcheck.
It is difficult when you try to get to a too-low level, but in general it should be possible to calculate some figure from population size, industrialisation level, expected growth over next decade, etc.
And when you get it wrong, it is always possible to relocate a (reserved but not yet used) block from one area to another, and have a situation that is not ideal but still better than it is now.
I think that within APNIC they should at least have done this. I.e. get some/8 blocks, use one in austrialia, one in china, part of one in korea, etc, and ask for more blocks when one gets depleted.
But no, they assign networks in the order the requests come in, often not even aligned on subnet boundaries! (i.e. you cannot describe an allocation as a single IP subnet)
Slashdot Mirror
Looking some more on the www.tulip.nl site, it turns out that it really is like I described before.
The current management have cut away everything that made Tulip a computer manufacturer, and only left a small bureau that defends some patents, trademarked brand names, and "markets".
A news article on the site shows a clear intent to go the SCO way: not design, sell or make any product, but just cash in on existing names and try to get money from others who are actually trying to do some innovation (and may unsuspectingly thread on something you have registered in the previous century).
The manager is proud of it, and enjoys the good life in a house in France, a fast car, and an office in a classic old city center.
Well, what should we say... it is a way to approach things, but I doubt if it will ever make them as successfull as e.g. Dell.
>The company has been on the brink of bankrupcy a number of times.
Probably again...
They show all the characteristics: digging through their old stack of patents and finding violations, and now looking in the pile of "brand names" they own and trying to cash-in on those.
From the entire article it is apparent that they expect nothing less than a steady stream of royalty money coming in all by itself by just declaring "commodore is our brand name", fighting all people who setup sites of their own, and bringing out some software emulator for the PC that they blindly assume 6 million people will buy from them.
I think it will be a great disappointment.
To patent something, you have to be a multinational, you must be greedy, you must think in terms of lawsuits, etc.
Farmers probably don't fit that bill, at least not originally.
components from SYSV...?
I bet they'll get a letter from SCO soon!
The pirates that abuse the Usenet system to transport their so-called "Warez" already have such a system in place.
The loss of part of the data can be overcome by extra (R-S) redundancy added to it when it was sent.
It is possible that this has already happened," he added with a weak smile, "but there is a certain amount of Uncertainty about it."
(Douglas Adams)
The reason they are doing this, is that the price level outside the US is usually higher.
Dell and Compaq sell systems all over the world, and when one would convert the US price plus shipping and tax to local currency, it would often be less than the local price from them.
So, to prevent people taking advantage of the world wide web and order their system from the US, they put some clauses into the contract to prevent you from doing that.
Of course this is protectionism, and in many countries it is outlawed. But they get away with it, apparently.
They should have asked a second question:
So do you plan to reduce your SCO Unix deployment, or eliminate SCO Unix entirely from your organization, as a result of SCO's threats?
The positive replies on that would probably be more than 9%...
No, that is not a correct representation.
A lot of things are illegal in Holland, but nobody ever bothers to do something about it.
You may know about cannabis posession. It isn't legal but no policeman or judge will bother.
What they did here is certainly illegal. However, they will most likely get away with it.
Even when the original owners of the domain speak up, they will most likely only hear that cases like this are not a priority of the legal system.
(the priorities of the legal system are very unclear. there is a definite bias towards cases that yield a good income, like putting up a camouflaged speed trap on a motorway behind a "30km because of roadworks" sign, or towing away cars from places where it not allowed to park them. finding and locking away the thief of your bicycle or laptop isn't a priority. hence, plain citizens are a victim but criminals rarely are)
>If you ask me (which nobody has), we should be using lower frequencies, not higher frequencies. Sure, lower frequencies require more bandwidth for the same speeds, but with lower bandwidths, obstacles wouldn't be a big deal, and you could transmit far, far further with less power as well.
Small wonder that nobody asks you....
We use Mozilla at work. I did almost all of the deployment stuff.
It is *completely* automatic. When logged on, PC's get the newest version of Mozilla (both Windows 95 and Windows 2000), and it is configured for the logged-on user completely automatically. The user never has to type a thing, except their mail password.
How is this done? Well, indeed you have to overcome some of the nasty features of the configuration manager. Using some tricks it is possible to get rid of the salt, and to store the Mozilla profile inside the User's roaming profile, so that they don't have to choose the profile when Mozilla is launched.
(every user gets their own setup of Mozilla, including mail address, addressbook, and all other preferences, as determined by their logon)
The first trick is to remove the profile, create a new profile named "default", change its directory to a location that is the same for each user (network drive, profile directory via a SUBST) and save that. Keep the registry.dat file that has been created in a safe place, and copy it over at every login. This resolves the salt problem and keeps your profile in the same place for every user.
To setup a new user, take the prefs.js of a standard user, delete all lines that refer to the user (mail address etc) and save that as a base config. Write a simple script that takes this file, appends the user-specific lines, and creates prefs.js. Then your users don't need to config anything.
For calendar, we use Maorong Zou's "WebCalendar", a web application running on a Linux server. It is not local to the browser, but it works well.
So far, it is all very nice. There are of course problems as well:
- Users complain it is slow. Especially the group of users that still has Pentium-class 64MB systems. It sure could be faster.
- There is a performance problem in the IMAP code. Downloading large attachments over the LAN is painfully slow, even on very fast systems. This is an open bug.
- There sometimes are problems with printing. Mozilla re-loads the page when printing it, sometimes several times. It often crashes when complicated javascript is in the page to be printed.
- There is an irritating problem when opening a new window, where it tends to re-use existing windows that contain valuable info (like your calendar). Workaround is to use tabs instead of windows.
Of course I can go on for a long time on the topic of using Mozilla, the advantages, the disadvantages, the tricks of installation and configuration, etc.
But overal I am quite happy. Maybe not all the users are, but that usually is a matter of "I am accustomed to Outlook and IE". Those people are usually also accustomed to viruses.
>It is a fairly trivial matter for most regular /. readers to back trace a spam mail to the source server. In nearly all cases the server is an open relay or has been owned - either way the plug should be pulled.
/. reader.
I think you have not looked at the matter last year.
What you say may have been true in the past, but the spammer's tactics have changed.
They use proxies now, not relays.
There is no way to trace the path back to them, for a regular
You would need co-operation from the access provider of an "innocent" family using cable or adsl internet, and from that family.
Well, in fact even from losers like the author of AnalogX Proxy and other Windows proxies that are by default open to the Internet and do not log.
>Is there any simple (as in Joe User simple, not simple as in run this script, patch this file, compile this kernal simple) way to get WinModem support under Linux?
How long until telephone modem support is just as low a priority as floppy support?
The floppy driver has been terrible for many years now, and nobody seems to care. Who uses floppies anyway?
I think it will go the same way with modems.
This is not a difference with Windows.
Most people cannot install Windows either. They get it installed on their new PC.
When they would have to install it, let it find hardware, and install drivers all from scratch they would find that daunting too.
It recurses until some pre-set depth
In this filter, executable files are not only recognized by their extension (and .msw would be very easily added to the list) but also by the first couple of bytes of the file.
All the processor-executable files have the same header structure in Windows, and are thus very easily identified.
With every next worm, I wonder why there does not appear one that first propagates and then erases all data it can touch.
You know, like the good old days when there was supposed to be a data in the near future when all PCs in the universe would crash because of a virus.
(quickly purchase a virus scanner or you will be doomed)
For that, you only need to be a Windows user.
The message seems to be coming from a friend, has an attachment that promises to be a document, when you unzip it it contains a file named like a document, so the normal next step would be to doubleclick on it, expecting it to be opened.
That this means "run it" in this case is a distinction that has been blurred by Windows.
My filter declines .zip files that contain executable files, but it passes .zip files that contain only documents.
Are you trying to say that not all filters would be capable of doing that?
> If you had bought
We have no choice what we buy. We get Windows as part of the workstation (the Microsoft Tax), it would be foolish to pay the tax and then buy another version and run that...
But it is good to hear that something is being done about it. With Linux, we don't have that problem.
(all the languages are on one set of CDs)
Actually it is a big shame that there are still language-dependencies in the OS, so that you need a different version of patches for each language.
(same for drivers etc)
We run English versions on the servers (for legacy reasons and because some things just aren't available in all languages), and have Dutch Windows 2000 on the workstations. Joy, joy, joy...
Probably we will have to wait a while for SP4.
>However, we do most definitely appreciate the effort put into submitting clean bug reports, and whenever possible try to follow-up on them.
I certainly do not want to criticize the tremendous work done by a lot of volunteers. When I did not have a lot of projects myself, I would probably be able to assist... but for this application, I am merely a consumer that is willing to report problems and spend some time describing a test case. Of course, it helps when knowing that actually something is going to improve because of these reports.
The two bugs I reported myself are 207607 and 207611.
Slow IMAP is reported in some different bugs, but the main one is 147285. Someone is working on that now, it seems.... but too late for 1.4 I fear.
>I really think that I should write a filter that spell-checks an email, and rejects it if over 50% of the words with 5 or more letters are misspelled.
That is a good idea. Of course it will need a large dictionary of correctly-spelled words in a couple of languages.
Maybe it is also possible to create an algorithm that decides if a string of characters is likely to be a valid word in a western language. I.e. that separates the correct words from the hwreqwuir. That could be used like the spellcheck.
Rob
It is difficult when you try to get to a too-low level, but in general it should be possible to calculate some figure from population size, industrialisation level, expected growth over next decade, etc.
/8 blocks, use one in austrialia, one in china, part of one in korea, etc, and ask for more blocks when one gets depleted.
And when you get it wrong, it is always possible to relocate a (reserved but not yet used) block from one area to another, and have a situation that is not ideal but still better than it is now.
I think that within APNIC they should at least have done this. I.e. get some
But no, they assign networks in the order the requests come in, often not even aligned on subnet boundaries! (i.e. you cannot describe an allocation as a single IP subnet)
I think that is ridiculous.