Last year I put together two new machines with nVIDIA graphics cards.
They both run Fedora, and I've had to get used to re-building the interface module for the (binary) graphics drivers each time a new kernel is released.
Yes, I know there are compiled modules available from livna, but I've had some problems with these in the past.
So far, +1 for nVIDIA and +1 for Linux.
However....
Next time I build another machine, I'll look at using a different graphics card that does not need manual intervention to update the graphics driver.
So, -1 for nVIDIA.
My non-tech brother has a similar machine, and problems with installing/updating the graphics drivers is one of the reasons why I haven't recommended he change to Linux... yet.
So, -1 for Linux.
Both of these would be solved if nVIDIA supported an Open Source graphics driver that RedHat could ship as part of the main distro (and would get updated automagically when new kernel is released).
So, we don't want the Chinese government doing things like snooping and blocking internet access, because this is a bad thing.
On the other hand, it is ok for the UK government to want a backdoor into peoples home computers because.... ?
I don't mean to belittle the efforts of the group in Canada, I think they are providing a useful service.
On the other hand, I'm wondering how long before people in other countries start offering secure file store services to people in the UK and the US ?
Perhaps we could trade... you give me file space on your computer to store my data, and I give you access to blocked internet sites.
Technically possible, but we would end up having to write (and test) our own custom URL connection handlers for each special case.
Our code uses 3rd party tools like Apache Axis to handle the SOAP connections.
The current version of Axis probably does use the system URL handlers, but I know that Axis 2.0 uses the Apache HttpClient tools instead.
If our code is shipped as a library, to be used as a plug-in for a 3rd party applications, they would not appreciate us hooking all 'http://' connections to use our own custom connector. They may have their own custom connector, and you can't chain the connectors together.
As far as I know, the only way to select IPV4 or IPV6 is to set a global system property. If the client is contacting many different servers, then it will constantly be trying to swap between IPV4 and IPV6 for each different connection.
Again, not an impossible thing to fix. However, the key point is that when developing business level logic, we shouldn't need to go hacking at the level of IP protocols.
If I use a JDBC connection to a database system, I should not need to worry about what kind of file system partition the database is mounted on.
Likewise, if I open a http connection to a web server, I should not need to worry about which version of IP protocol the network routers use.
For IPV6 to be adopted, these sort of things should be handled in the JVM layer. Not in the application code.
Many consumers use NAT because their ISP would only give them one IP address.
That's not going to change going from IPv4 to IPv6 without other structural changes at the ISPs.
They're just not set up to track multiple addresses for one customer, no matter what kind of addresses those are.
Their databases only have one field for that, and that one field gets used to seed their MAC filter etc.
Where's the incentive for them to accomodate this guy's wishes?
From their web site :
Our ADSL service is IPv6 enabled; all customers will get a single static IPv6 address allocated to them by default if their equipment can support it.
We can also allocate a/48 of IPv6 address space if a customer wishes to deploy it on their internal network.
In light of recent coverage, I think they need to update the press page on their website.
http://www.xcp-aurora.com/press_related.aspx/
Methinks they might need to dedicate a whole page just for Oct 2005 alone;-)
Current Java support for IPV6 is broken.
Not in a technical sense, everything works fine technically.
Java detects if the local host OS supports IPV6 and will try to use it when available,
which seems like a GoodThing(TM).
When you ask the URLConnection class to open a connection Java does two DNS lookups,
one for the IPV4 'A' record, and a second one for an IPV6 'AAAA' record.
If the DNS server replies with a valid 'AAAA' record, then Java will try to use
IPV6 to connect to the remote host, which also seems like a GoodThing(TM).
However, checking for a valid 'AAAA' DNS record only tells you if the remote system supports IPV6.
It does not tell you anything about the nodes in between here and there.
We are working on a Grid system using SOAP as the core message protocol.
Most of our servers are hosted at university departments, and most of them already
run IPV6 networks and will have been issued with valid 'AAAA' records.
However, we want users to be able to connect to our systems from outside.
In fact, most of the development team work from home a lot of the time.
If I run a Java app on a Linux machine at home and try to connect to a server
at a university:
Java detects that the OS (Linux) does indeed support IPV6
Java detects that the remote host has a valid 'AAAA' record
Java assumes that everyone can handle IPV6 and decides to use it
However:
The ADSL router connecting my network to the outside world does not support IPV6
The ADSL line provided by my ISP does not support IPV6
All of the packets evaporate into the ether
The Java URLConnection gives up and throws a 'Network is unreachable' exception
The URLConnection does not re-try using IPV4
The last one is the problem.
If it caught the Exception, and then tried again with IPV4, then we could use IPV6 where possible, and still fall back to IPV4 if needed.
As far as I can tell, there isn't a way of selecting IPV4/IPV6 on a per connection basis.
The only way to select it is using a global system property at startup.
Which means that in order to support the edge case of professor using our system on laptop
from home, we have to ship all our software with IPV6 dispabled.
If anyone knows of a fix for this please let me know.
My eyes gravitate towards whatever article/information I'm reading and completely ignores the peripheral ads
The text based ads down the side and banner ads accross the top are fine.
I do remember a database vendor ad on slashdot a few months ago, which had a large animated monster bashing someone over the head.
This particular ad was placed in the main page, and was almost impossible to ignore.
I found this ad so annoying and distracting, I stopped visiting slashdot for a while and it made me hostile to even considering using their product.
By us not clicking on the crappy flash ads -- that sends a message. Blocking it does not
If they want feedback, they (the advertising companies) could work with the FireFox team to provide a 'block this particular advert' option.
I wouldn't mind if the browser sent a message back to the advertising site to tell them that I had blocked their advert.
Effectively telling them 'I have seen your advert, decided it is annoying, and have blocked it'.
They would get accurate feedback on which adverts get blocked, and we (hopefully) would see fewer annoying ones.
Ok, I know this would not work in the real world... but it would be nice.
It is also not that hard to write short paragraph that conforms to the rules of English grammar.
This is not true. It is not really fucking hard. You just have
to have sufficient time and not be so lazy. It
may take 4 times what you normally think, and be boring as
all hell, but its totally possible.
Note - I'm not complaining about the grammar, or lack of, in your post. I am just pointing out that you seem to have considered what you had 'good enough' to get the point across.
I agree with you, it is possible to check all of the details are correct, it just takes time and attention to detail.
I'd have thought that clear written details of the transaction would be important for high-end goods.
If I write, in text, that the car is roadworthy and has all the relevant certificates, then there is an audit trail published on the web site.
If I say, in a voice call, that the car is roadworthy, where is the audit trail ?
From the article :
.... eBay is also attracted by the idea of letting its buyers and sellers talk to each other via their computers...
Personally, I'm not sure that I would want eBay buyers and sellers to contact me by voice.
For a start, it means that to buy or sell effectively, you would need to be online a lot of the time.
Plus, once they have your Skype address, it would open up the system to SPAM voice calls
pestering you to buy more things that you don't want or need.
Not all terrorist attacks are suicide attacks. Maybe you havn't been paying attention to the news lately, but not too long ago the London subways were bombed and surveillance cameras helped police determine their identities.
The cameras are primarily used for monitoring criminal rather than terrorist activity.
http://news.independent.co.uk/uk/transport/article 307679.ece A spokesman said that the police made regular requests to see footage as part of their investigations, although the vast bulk of such requests involved criminal rather than terrorist activity.
When the cameras could have caught images of a suspected terrorist... none of them were working.
http://news.independent.co.uk/uk/crime/article3076 49.ece None of the cameras at the scene of the shooting of Jean Charles de Menezes at Stockwell Tube station on 22 July were working, a police document revealed.
This is what makes C difficult for beginners. Bad code compiles even though it has logical flaws, and ends up failing in mysterious ways.
A better way of writing the same thing is
if (NULL == ptr)
I know that many 'hackers' will object because
a) "It is more characters to type"
and
b) "It reads the wrong way round and bugs me"
However, there is a reason in my madness....
The difference between
if (!ptr)
and
if (ptr)
Is a one character typo, both compile, but but the second one does not do what you expect.
The difference between
if (ptr == NULL)
and
if (ptr = NULL)
Is also a one character typo, both compile, but the second one does not do what you expect.
In fact, the second one changes the value of ptr, making it very hard to find the cause when you try to use ptr 100 lines later.
The difference between
if (NULL == ptr)
and
if (NULL = ptr)
Is a compiler error for the second form, you can't assign a new value to '0'.
If you are comparing a variable with a constant, put the constant first.
Cole :
Did you underestimate the value of security?
Gates :
Certainly you can never underestimate the level of malicious people out there who are going to try to take advantage of whatever things there are.
Me :
Microsoft (and possibly many others in the IT industry) did underestimate the threat.
The threat they missed was not from 'malicious people' but the ease with which viruses and worms could spread in an interconnected world.
Gates :
That's why we made trustworthy computing the top priority.
Me :
This was only after Open Source alternatives began to gain market share because they were perceived as being less vulnerable.
Is 'trustworthy computing' really addressing the problem of multiple security weaknesses in the operating system ?
Or, is it a way to make it more difficult for Open Source software to gain market share by only allowing the user to install
software from 'trusted' (commercial) suppliers.
Cole :
Nevertheless, a lot of our viewers still say to us: 'Microsoft didn't take that threat seriously enough and we are having problems.'
Gates :
Certainly we can always do better. It's the top priority. If you look at things like spam, we feel very good about the progress there.
Me :
The question was about security.
Spam is irritating, but it is not a security problem.
Viruses, worms and identity theft are security problems.
Spam is only a security problem when a virus or worm turns an innocent home computer into a zombie spam robot.
They would have to have very very good thermal insulation.
... the fluid involved is methane, a simple organic compound that can exist as a liquid or gas at Titan's sub-170C temperatures..
From what I have heard, this is one of the reasons the probe had such a short lifespan, batteries don't last long at these kind of temperatures.
I suspect that this would make even a rover type robot quite a difficult challenge.
The Huygens team held a press conference this morning and presented some of the results of their analysis so far.
The first scientific assessments of Huygens' data were presented during a press conference at ESA head office in Paris on 21 January.
Results include:
Geological evidence for precipitation, erosion, mechanical abrasion and other fluvial activity says that the physical processes shaping Titan are much the same as those shaping Earth
Huygens' data provide strong evidence for liquids flowing on Titan. However, the fluid involved is methane.
... while many of Earth's familiar geophysical processes occur on Titan, the chemistry involved is quite different. Instead of liquid water, Titan has liquid methane. Instead of silicate rocks, Titan has frozen water ice. Instead of dirt, Titan has hydrocarbon particles settling out of the atmosphere, and instead of lava, Titanian volcanoes spew very cold ice...
He'd be a lot more believable if.... he just..shaved
Why ?
Do you really belive that clean shaven people are more trustworthy ?
My experience is that 'odd looking' people may have some odd ideas, but they tend to really believe in them.
Sharp looking clean shaven people in business suit and tie are often using the illusion of respectability to hide their real intentions.
People who think outside the box often ignore the social codes and dress in a way that they feel comfortable.
Often this is to filter out people they don't want to bother with.
It works the same way as the business suit, but selects a different group of people.
But he did have time to add this to the end of his page.
<!-- Start of StatCounter Code --> <script type="text/javascript" language="javascript"> var sc_project=465856; var sc_partition=2; </script>
<script type="text/javascript" language="javascript" src="http://www.statcounter.com/counter/counter.js "></script><noscript><a href="http://www.statcounter.com/" target="_blank"><img src="http://c3.statcounter.com/counter.php?sc_proj ect=465856&java=0" alt="frontpage hit counter" border="0"></a> </noscript> <!-- End of StatCounter Code -->
From the http://www.statcounter.com/ site :
"A free yet reliable invisible web tracker, highly configurable hit counter and real-time detailed web stats."
It is on Slashdot for the same reason it is on the BBC news.
It is something that we should all remember.
Many of us work in technology related jobs, and it is important for us to remember the mistakes of the past in order to avoid them in the future.
We all have a responsibility to make sure something like this does not happen again.
In the work we do, in our choices of things we buy, and in legislation our respective government(s) apply.
There may be someone reading this who is worried about the safety procedures where they work.
This reminder may help them to make the right decision.
When we buy things because they are cheaper, do we ask how a company can afford to sell something at
significantly lower cost than others ?
If we don't ask, or don't care, then we may be (indirectly) encouraging a less safe manufacturing process.
Companies won't take the risks if we don't buy the products.
In the UK, the government is working on a bill to introduce a new offence of 'Corporate Killing'.
http://www.corporateaccountability.org/Updates/man slaughter.htm#current
Would this help to make companies pay more attention to possible safety risks ?
We should all encourage our respective governments to improve their legislation.
I'm as guilty as the next person.
I've seen things in the past that are not 100% safe, and have not reported them.
I buy things that are cheaper and haven't checked how they were made.
I haven't done much to push for better legislation.
Perhaps this article is a reminder for us to do better... ?
I'm glad I'm not the only one:)
Whenever I upgrade to a new machine, I throw away the shiny new 'soft touch' keyboard and plug in my trusty old IBM original.
I don't know what I'll do when new machines no longer support a PS2 keyboard....
Slashdot Mirror
Last year I put together two new machines with nVIDIA graphics cards.
They both run Fedora, and I've had to get used to re-building the interface module for the (binary) graphics drivers each time a new kernel is released.
Yes, I know there are compiled modules available from livna, but I've had some problems with these in the past.
So far, +1 for nVIDIA and +1 for Linux.
However ....
So, -1 for nVIDIA.
So, -1 for Linux.
Both of these would be solved if nVIDIA supported an Open Source graphics driver that RedHat could ship as part of the main distro (and would get updated automagically when new kernel is released).
I work as a software developer for a department at Cambridge.
We are part of a distributed project, with team members in other institutes within the UK and around the world.
We use both Skype and Jabber to collaborate with each other.
So, we don't want the Chinese government doing things like snooping and blocking internet access, because this is a bad thing. .... ?
... you give me file space on your computer to store my data, and I give you access to blocked internet sites.
On the other hand, it is ok for the UK government to want a backdoor into peoples home computers because
I don't mean to belittle the efforts of the group in Canada, I think they are providing a useful service.
On the other hand, I'm wondering how long before people in other countries start offering secure file store services to people in the UK and the US ?
Perhaps we could trade
Technically possible, but we would end up having to write (and test) our own custom URL connection handlers for each special case. Our code uses 3rd party tools like Apache Axis to handle the SOAP connections. The current version of Axis probably does use the system URL handlers, but I know that Axis 2.0 uses the Apache HttpClient tools instead.
If our code is shipped as a library, to be used as a plug-in for a 3rd party applications, they would not appreciate us hooking all 'http://' connections to use our own custom connector. They may have their own custom connector, and you can't chain the connectors together.
As far as I know, the only way to select IPV4 or IPV6 is to set a global system property. If the client is contacting many different servers, then it will constantly be trying to swap between IPV4 and IPV6 for each different connection.
Again, not an impossible thing to fix. However, the key point is that when developing business level logic, we shouldn't need to go hacking at the level of IP protocols.
For IPV6 to be adopted, these sort of things should be handled in the JVM layer. Not in the application code.
Many consumers use NAT because their ISP would only give them one IP address. That's not going to change going from IPv4 to IPv6 without other structural changes at the ISPs. They're just not set up to track multiple addresses for one customer, no matter what kind of addresses those are. Their databases only have one field for that, and that one field gets used to seed their MAC filter etc. Where's the incentive for them to accomodate this guy's wishes?
Black Cat Networks
http://www.blackcatnetworks.co.uk/services/adsl
From their web site : /48 of IPv6 address space if a customer wishes to deploy it on their internal network.
Our ADSL service is IPv6 enabled; all customers will get a single static IPv6 address allocated to them by default if their equipment can support it. We can also allocate a
Sorry, typo in link :-(
http://www.xcp-aurora.com/press_related.aspx
In light of recent coverage, I think they need to update the press page on their website. ;-)
http://www.xcp-aurora.com/press_related.aspx/
Methinks they might need to dedicate a whole page just for Oct 2005 alone
Current Java support for IPV6 is broken. Not in a technical sense, everything works fine technically.
Java detects if the local host OS supports IPV6 and will try to use it when available, which seems like a GoodThing(TM).
When you ask the URLConnection class to open a connection Java does two DNS lookups, one for the IPV4 'A' record, and a second one for an IPV6 'AAAA' record.
If the DNS server replies with a valid 'AAAA' record, then Java will try to use IPV6 to connect to the remote host, which also seems like a GoodThing(TM).
However, checking for a valid 'AAAA' DNS record only tells you if the remote system supports IPV6. It does not tell you anything about the nodes in between here and there.
We are working on a Grid system using SOAP as the core message protocol. Most of our servers are hosted at university departments, and most of them already run IPV6 networks and will have been issued with valid 'AAAA' records.
However, we want users to be able to connect to our systems from outside. In fact, most of the development team work from home a lot of the time.
If I run a Java app on a Linux machine at home and try to connect to a server at a university:
However:
The last one is the problem. If it caught the Exception, and then tried again with IPV4, then we could use IPV6 where possible, and still fall back to IPV4 if needed.
As far as I can tell, there isn't a way of selecting IPV4/IPV6 on a per connection basis. The only way to select it is using a global system property at startup.
Which means that in order to support the edge case of professor using our system on laptop from home, we have to ship all our software with IPV6 dispabled.
If anyone knows of a fix for this please let me know.
My eyes gravitate towards whatever article/information I'm reading and completely ignores the peripheral ads
The text based ads down the side and banner ads accross the top are fine. I do remember a database vendor ad on slashdot a few months ago, which had a large animated monster bashing someone over the head.
This particular ad was placed in the main page, and was almost impossible to ignore. I found this ad so annoying and distracting, I stopped visiting slashdot for a while and it made me hostile to even considering using their product.
By us not clicking on the crappy flash ads -- that sends a message. Blocking it does not
If they want feedback, they (the advertising companies) could work with the FireFox team to provide a 'block this particular advert' option.
I wouldn't mind if the browser sent a message back to the advertising site to tell them that I had blocked their advert. Effectively telling them 'I have seen your advert, decided it is annoying, and have blocked it'.
They would get accurate feedback on which adverts get blocked, and we (hopefully) would see fewer annoying ones.
Ok, I know this would not work in the real world ... but it would be nice.
It is also not that hard to write short paragraph that conforms to the rules of English grammar.
This is not true. It is not really fucking hard. You just have to have sufficient time and not be so lazy. It may take 4 times what you normally think, and be boring as all hell, but its totally possible.
Note - I'm not complaining about the grammar, or lack of, in your post. I am just pointing out that you seem to have considered what you had 'good enough' to get the point across.
I agree with you, it is possible to check all of the details are correct, it just takes time and attention to detail.
While looking through the documentation http://www.archives.gov/era/about/documentation.h
I found a link to the project requirements : http://www.archives.gov/era/about/requirements.cs
Which contains the following line
I know, one typo in one line in several hundred, but why that line ?
I'd have thought that clear written details of the transaction would be important for high-end goods.
If I write, in text, that the car is roadworthy and has all the relevant certificates, then there is an audit trail published on the web site. If I say, in a voice call, that the car is roadworthy, where is the audit trail ?
From the article :
.... eBay is also attracted by the idea of letting its buyers and sellers talk to each other via their computers ...
Personally, I'm not sure that I would want eBay buyers and sellers to contact me by voice. For a start, it means that to buy or sell effectively, you would need to be online a lot of the time.
Plus, once they have your Skype address, it would open up the system to SPAM voice calls pestering you to buy more things that you don't want or need.
I wonder how much of it is covered by software patents ?
Not all terrorist attacks are suicide attacks. Maybe you havn't been paying attention to the news lately, but not too long ago the London subways were bombed and surveillance cameras helped police determine their identities.
The cameras are primarily used for monitoring criminal rather than terrorist activity.e 307679.ece
http://news.independent.co.uk/uk/transport/articl
A spokesman said that the police made regular requests to see footage as part of their investigations, although the vast bulk of such requests involved criminal rather than terrorist activity.
When the cameras could have caught images of a suspected terrorist ... none of them were working.
6 49.ece
http://news.independent.co.uk/uk/crime/article307
None of the cameras at the scene of the shooting of Jean Charles de Menezes at Stockwell Tube station on 22 July were working, a police document revealed.
A better way of writing the same thing is
I know that many 'hackers' will object because
a) "It is more characters to type"
and
b) "It reads the wrong way round and bugs me"
However, there is a reason in my madness
The difference betweenandIs a one character typo, both compile, but but the second one does not do what you expect.
The difference betweenandIs also a one character typo, both compile, but the second one does not do what you expect. In fact, the second one changes the value of ptr, making it very hard to find the cause when you try to use ptr 100 lines later.
The difference betweenandIs a compiler error for the second form, you can't assign a new value to '0'.
If you are comparing a variable with a constant, put the constant first.
Cole :
Did you underestimate the value of security?
Gates :
Certainly you can never underestimate the level of malicious people out there who are going to try to take advantage of whatever things there are.
Me :
Microsoft (and possibly many others in the IT industry) did underestimate the threat.
The threat they missed was not from 'malicious people' but the ease with which viruses and worms could spread in an interconnected world.
Gates :
That's why we made trustworthy computing the top priority.
Me :
This was only after Open Source alternatives began to gain market share because they were perceived as being less vulnerable.
Is 'trustworthy computing' really addressing the problem of multiple security weaknesses in the operating system ?
Or, is it a way to make it more difficult for Open Source software to gain market share by only allowing the user to install software from 'trusted' (commercial) suppliers.
Cole :
Nevertheless, a lot of our viewers still say to us: 'Microsoft didn't take that threat seriously enough and we are having problems.'
Gates :
Certainly we can always do better. It's the top priority. If you look at things like spam, we feel very good about the progress there.
Me :
The question was about security.
Spam is irritating, but it is not a security problem.
Viruses, worms and identity theft are security problems.
Spam is only a security problem when a virus or worm turns an innocent home computer into a zombie spam robot.
The new version reads as follows :
They would have to have very very good thermal insulation.
... the fluid involved is methane, a simple organic compound that can exist as a liquid or gas at Titan's sub-170C temperatures ..
From what I have heard, this is one of the reasons the probe had such a short lifespan, batteries don't last long at these kind of temperatures.
I suspect that this would make even a rover type robot quite a difficult challenge.
The first scientific assessments of Huygens' data were presented during a press conference at ESA head office in Paris on 21 January.
Results include:
He'd be a lot more believable if .... he just..shaved
Why ?
Do you really belive that clean shaven people are more trustworthy ?
My experience is that 'odd looking' people may have some odd ideas, but they tend to really believe in them.
Sharp looking clean shaven people in business suit and tie are often using the illusion of respectability to hide their real intentions.
People who think outside the box often ignore the social codes and dress in a way that they feel comfortable.
Often this is to filter out people they don't want to bother with.
It works the same way as the business suit, but selects a different group of people.
From the http://www.statcounter.com/ site : "A free yet reliable invisible web tracker, highly configurable hit counter and real-time detailed web stats."
It is on Slashdot for the same reason it is on the BBC news.
n slaughter.htm#current
... ?
It is something that we should all remember.
Many of us work in technology related jobs, and it is important for us to remember the mistakes of the past in order to avoid them in the future.
We all have a responsibility to make sure something like this does not happen again.
In the work we do, in our choices of things we buy, and in legislation our respective government(s) apply.
There may be someone reading this who is worried about the safety procedures where they work. This reminder may help them to make the right decision.
When we buy things because they are cheaper, do we ask how a company can afford to sell something at significantly lower cost than others ?
If we don't ask, or don't care, then we may be (indirectly) encouraging a less safe manufacturing process.
Companies won't take the risks if we don't buy the products.
In the UK, the government is working on a bill to introduce a new offence of 'Corporate Killing'. http://www.corporateaccountability.org/Updates/ma
Would this help to make companies pay more attention to possible safety risks ?
We should all encourage our respective governments to improve their legislation.
I'm as guilty as the next person.
I've seen things in the past that are not 100% safe, and have not reported them.
I buy things that are cheaper and haven't checked how they were made.
I haven't done much to push for better legislation.
Perhaps this article is a reminder for us to do better
Your opinion causes me great sadness.
I'm glad I'm not the only one :)
....
Whenever I upgrade to a new machine, I throw away the shiny new 'soft touch' keyboard and plug in my trusty old IBM original.
I don't know what I'll do when new machines no longer support a PS2 keyboard