They didn't have to screw the kernel, their waagent is a hideous piece of crap with can do that in userspace. When I glanced through it a while back it was doing VERY stupid things like creating a swap file with perm mode 666. Yes. I think this was one of the commits https://github.com/Azure/WALin... - if not look through the history. How stuff like that gets out the door I have no idea. What worries me is that within MS they have project ideas, but not the Unix Sysadmin skills to know what they're doing is wrong, then blaming the community for not submitting PRs to fix their issues. Yes, open source works though that for GPL ideas, but at the same time I feel MS are abusing it somewhat as free labour.
I could live without social media if I could keep in touch with family and friends and also find a place to talk about stuff with anyone who's interested.
If you can make IRC really user friendly so that the less technically inclined can use it and reincarnate usenet, it would be a very small start. In 20 years, we might have something.
I think AOL did this a while back but too many rowdy teenagers were cluttering the place.
So what? the costs get passed on anyway, if it is more expensive to gather then it just becomes a more valuable commodity.
For it to have value then it must have importance, worth, or usefulness of something. Not just be expensive. Fragmenting social network sites could put up the aggregation cost, then perhaps the data becomes less useful and more complex to merge. Given the current state of things I'm willing to try something different.
On the other hand, would you pay 10GBP more for an item where the social media data mining activity cost has been merged into the shelf price? I'd go for the one where I'm not paying that tax, given the choice of course.
The use case is that someone has a need to deliver content over FTP.
Your logical fallacy is: Circular Argument! Now try to come up with a use case for delivering content over FTP within a HTTP site without that use case being "But someone may need to!!!"
You're misusing the logical fallacy argument, we're talking established protocols here, not adding something that wasn't there already. The use case is, as said already, FTP server is likely to have less overheads. Take a look at RHEL Apache installs as an example, they're the kitchen sink, unlike that vsftp. Take this to an extreme with a dynamic site that handles all associated mod_* hooks and you will very likely have much slower delivery than a skinny vsftp.
doesn't remove the fact that someone may wish to deliver that way
I thought you were talking about a use case. If someone "may wish to" without a use case, all the while delivering a poor solution, then I will continue to call them out for the morons they are.
No browsers should not continue to appease people who misuse their features stupidly.
The only misuse you came up with was marginal performance. It is nonsensical to reduce protocol coverage based on that alone. There is still plenty of room in the world for mixed protocol delivery.
I think you may not have read my reply if you think for a moment that your post contained a "use case". No your post contained a clunky way of doing something that was never intended to do without any benefit and a load of downsides. That does not a use case make, and you have yet to come up with an actual use case for serving inline HTTP content via FTP.
Not only is there no need for it, EVER, it also is a stupid idea if you sadistically have a *want* for it.
The use case is that someone has a need to deliver content over FTP. That's the long and short of it. Just because you don't want to do that doesn't remove the fact that someone may wish to deliver that way. You may choose to not do that if you wish, once the browsers include a way to do that they should continue to include that otherwise they have removed functionality that may leave some users surprised. Regardless of you choice delivery protocol preferences.
Again, you have totally missed the point. There are use cases but you refuse to look at other needs. Go back and read my original post, HTTP isn't the only way to transfer content.
Anybody web designer who considers this "functionality" for serving mid page content should have all their fingers broken so they feel some proper pain when they type.
Really? There was a time when it was not as simple to have HTTP servers everywhere. There could still be situations where FTP could be faster. Many HTTP servers require more memory than FTP, and, all things being equal there could be real situations where an FTP may be faster than an HTTP, badly written mod_perl/mod_php that sits at the header processing stage could make HTTP perform worse than FTP. Maybe just to take load away from the HTTP you might decide to put that traffic through FTP if its static. This could be quite rare, and you wouldn't be going through all the faff that a normal FTP client goes through on login since you know ahead of time where the content is so you'd likely PASV, BINARY then GET. Yes, would likely be much slower since these would be synchronous but I can see an edge case where a "developer" may be in a bind.
This shouldn't, and if the developers are competent, wouldn't affect many people at all.
This makes no sense to me whatsoever. I fear there is a greater quantity of exploited HTTP(S) servers out there than FTP. Is this not akin to removing telnet from Windows? The loss of functionality does not match the gain in security (is there any?). Surely the first step should be to prevent malicious content, not prevent a protocol.
Are Mozilla thinking to block FTPS too? What about sftp (if it were ever to be introduced), would that count too?
If the argument is that the protocol is plaintext, then HTTP should be dropped.
And FTP is hard for search engines to index with useful metadata for advertisers.
I think it is somewhat easier for a search engine to 'list -al' the directory to see which timestamps have changed compared to performing a HEAD on each request, parsing the content and spidering. If a robots.txt exists, great, is is accurate? Dunno, will have to index anyway and see.
It's a testament to the backward compatibility of Windows itself, especially that this was originally included in Windows more than 20 years ago
Things like opendir, readdir, closedir haven't changed a great deal. $company opensources some abandonware to get some type of media attention, why's this news?
The difference is the setup. A Unix-like OS will segregate by user and root is rarely needed. In a MS OS you need Administrator to do just about anything, and rarely is software standalone. Can you install Office without Administrator?
Typically, a service that you expose on the network will not have perms to write to the service storage area. So when malware comes in through the front door, it can't do much other than read storage. If there is an elevation path, well, game over. What's the chance of a service in a MS setup having a route through to admin via AD, just from the way it has been configured with AD access group rights scattered to the four winds. These days AD seems the Achilles heal of MS, not it's saviour.
Why would you use a VPN to do anything these days? The VPN offers nothing that a cheap throw away OpenVZ machine could, but at least with the OpenVZ you have some chance of knowing what the outbound traffic is doing, its much less likely that the provider is going to sniff you traffic.
Given there are more uses for a VPS than a VPN, I don't see the VPN market growing, this cannot be a good business move.
evilwm http://www.usenix.org.uk/conte... and/or ratpoison. Both provide great usability. If you spend all day in a GUI or text environment, you really should learn these.
People are brand-loyal and seem incapable of 'choosing not to buy.'
Me too! I/am/ brand-loyal. Loyal to the trusty 3.5mm I know so well. The little connector I've soldered many times to splitters, converters, and sometimes to weird things.
There is no argument for this, the manufacturers are in cahoots, they've formed a racket
The king of ever changing non standard connectors that have "premium priced" cables, dongles, etc. Do you see a pattern yet?
Yes, I get the bit about how they want to make money, but seriously? Who in the industry thinks the population is dumb enough to buy it.
I can't see a point in changing it, there's no technical benefit. Anything that goes wireless is going to need batteries, fair enough, you have me there, improvements in latency and databus can happen, but in the cabled variety, really?
Can't the public push back and choose to not buy? What can top a set of Sennheiser HD 206's? Anything above that price range isn't going to get any real, material benefit. Digital on the other hand is going to be subject to more variables and I don't honestly think people should be convinced they're getting any improvement there. Anything analogue that isn't 3.5mm are only buying something that has very limited use by comparison. Please world, make this nonsense stop.
It'd be nice if it could fall back to a last known good config like on Windows (not that that ever works, but the way Ubuntu seems to keep old ones seems like it could be made to work).
Exactly. The headline is rubbish. Part of the kernel installation is to leave the last one in the menu. It's just the new one is a default. If grub is configured with 'savedefault', then the last picked kernel will be chosen for future boots.
What you're saying in the extreme is that things will be free because there are zero costs, but that's not true, you'll still have to pay rates, taxes and materials. Problem is, most of the country will be unemployed so you'll not have a customer, even if things are free for them as they're too busy trying to keep a roof over their heads because, like you, they'll have no income. Zero customers, zero income.
Slashdot Mirror
Shame it's not the DB format used on the later iPod Nano music players, because that would have been useful.
They didn't have to screw the kernel, their waagent is a hideous piece of crap with can do that in userspace. When I glanced through it a while back it was doing VERY stupid things like creating a swap file with perm mode 666. Yes. I think this was one of the commits https://github.com/Azure/WALin... - if not look through the history. How stuff like that gets out the door I have no idea. What worries me is that within MS they have project ideas, but not the Unix Sysadmin skills to know what they're doing is wrong, then blaming the community for not submitting PRs to fix their issues. Yes, open source works though that for GPL ideas, but at the same time I feel MS are abusing it somewhat as free labour.
Typical conversation at a 2025 back-yard BBQ:
Sounds about write. In 2025 we'll be so dumb we misspell words when talking.
If you haven't done so already, watch https://www.imdb.com/title/tt0...
I could live without social media if I could keep in touch with family and friends and also find a place to talk about stuff with anyone who's interested.
If you can make IRC really user friendly so that the less technically inclined can use it and reincarnate usenet, it would be a very small start. In 20 years, we might have something.
I think AOL did this a while back but too many rowdy teenagers were cluttering the place.
> Is It Time To Stop Using Social Media?
When was it ever time to *start*?
It's an anti-user business model by design.
End of discussion.
I think what's happened is some businesses feel they can promote their wares or marketing through a captive audience.
So what? the costs get passed on anyway, if it is more expensive to gather then it just becomes a more valuable commodity.
For it to have value then it must have importance, worth, or usefulness of something. Not just be expensive. Fragmenting social network sites could put up the aggregation cost, then perhaps the data becomes less useful and more complex to merge. Given the current state of things I'm willing to try something different.
On the other hand, would you pay 10GBP more for an item where the social media data mining activity cost has been merged into the shelf price? I'd go for the one where I'm not paying that tax, given the choice of course.
The use case is that someone has a need to deliver content over FTP.
Your logical fallacy is: Circular Argument! Now try to come up with a use case for delivering content over FTP within a HTTP site without that use case being "But someone may need to!!!"
You're misusing the logical fallacy argument, we're talking established protocols here, not adding something that wasn't there already. The use case is, as said already, FTP server is likely to have less overheads. Take a look at RHEL Apache installs as an example, they're the kitchen sink, unlike that vsftp. Take this to an extreme with a dynamic site that handles all associated mod_* hooks and you will very likely have much slower delivery than a skinny vsftp.
doesn't remove the fact that someone may wish to deliver that way
I thought you were talking about a use case. If someone "may wish to" without a use case, all the while delivering a poor solution, then I will continue to call them out for the morons they are.
No browsers should not continue to appease people who misuse their features stupidly.
The only misuse you came up with was marginal performance. It is nonsensical to reduce protocol coverage based on that alone. There is still plenty of room in the world for mixed protocol delivery.
I think you may not have read my reply if you think for a moment that your post contained a "use case". No your post contained a clunky way of doing something that was never intended to do without any benefit and a load of downsides. That does not a use case make, and you have yet to come up with an actual use case for serving inline HTTP content via FTP.
Not only is there no need for it, EVER, it also is a stupid idea if you sadistically have a *want* for it.
The use case is that someone has a need to deliver content over FTP. That's the long and short of it. Just because you don't want to do that doesn't remove the fact that someone may wish to deliver that way. You may choose to not do that if you wish, once the browsers include a way to do that they should continue to include that otherwise they have removed functionality that may leave some users surprised. Regardless of you choice delivery protocol preferences.
Again, you have totally missed the point. There are use cases but you refuse to look at other needs. Go back and read my original post, HTTP isn't the only way to transfer content.
You seem to have spent a lot of energy confusing best solution with possible solution.
Anybody web designer who considers this "functionality" for serving mid page content should have all their fingers broken so they feel some proper pain when they type.
Really? There was a time when it was not as simple to have HTTP servers everywhere. There could still be situations where FTP could be faster. Many HTTP servers require more memory than FTP, and, all things being equal there could be real situations where an FTP may be faster than an HTTP, badly written mod_perl/mod_php that sits at the header processing stage could make HTTP perform worse than FTP. Maybe just to take load away from the HTTP you might decide to put that traffic through FTP if its static. This could be quite rare, and you wouldn't be going through all the faff that a normal FTP client goes through on login since you know ahead of time where the content is so you'd likely PASV, BINARY then GET. Yes, would likely be much slower since these would be synchronous but I can see an edge case where a "developer" may be in a bind.
This shouldn't, and if the developers are competent, wouldn't affect many people at all.
I don't think there are many descent web designers.
This makes no sense to me whatsoever. I fear there is a greater quantity of exploited HTTP(S) servers out there than FTP. Is this not akin to removing telnet from Windows? The loss of functionality does not match the gain in security (is there any?). Surely the first step should be to prevent malicious content, not prevent a protocol.
Are Mozilla thinking to block FTPS too? What about sftp (if it were ever to be introduced), would that count too?
If the argument is that the protocol is plaintext, then HTTP should be dropped.
And FTP is hard for search engines to index with useful metadata for advertisers.
I think it is somewhat easier for a search engine to 'list -al' the directory to see which timestamps have changed compared to performing a HEAD on each request, parsing the content and spidering. If a robots.txt exists, great, is is accurate? Dunno, will have to index anyway and see.
It's a testament to the backward compatibility of Windows itself, especially that this was originally included in Windows more than 20 years ago
Things like opendir, readdir, closedir haven't changed a great deal. $company opensources some abandonware to get some type of media attention, why's this news?
The difference is the setup. A Unix-like OS will segregate by user and root is rarely needed. In a MS OS you need Administrator to do just about anything, and rarely is software standalone. Can you install Office without Administrator?
Typically, a service that you expose on the network will not have perms to write to the service storage area. So when malware comes in through the front door, it can't do much other than read storage. If there is an elevation path, well, game over. What's the chance of a service in a MS setup having a route through to admin via AD, just from the way it has been configured with AD access group rights scattered to the four winds. These days AD seems the Achilles heal of MS, not it's saviour.
This must be pretty good from the anti-terrorism point of view if you're trying to work out who is a member of a terror network.
Why would you use a VPN to do anything these days? The VPN offers nothing that a cheap throw away OpenVZ machine could, but at least with the OpenVZ you have some chance of knowing what the outbound traffic is doing, its much less likely that the provider is going to sniff you traffic.
Given there are more uses for a VPS than a VPN, I don't see the VPN market growing, this cannot be a good business move.
evilwm http://www.usenix.org.uk/conte... and/or ratpoison. Both provide great usability. If you spend all day in a GUI or text environment, you really should learn these.
Tonight, on Cop Drama, they're going to say "shit" ... Twice!
People are brand-loyal and seem incapable of 'choosing not to buy.'
Me too! I /am/ brand-loyal. Loyal to the trusty 3.5mm I know so well. The little connector I've soldered many times to splitters, converters, and sometimes to weird things.
There is no argument for this, the manufacturers are in cahoots, they've formed a racket
The king of ever changing non standard connectors that have "premium priced" cables, dongles, etc. Do you see a pattern yet?
Yes, I get the bit about how they want to make money, but seriously? Who in the industry thinks the population is dumb enough to buy it.
I can't see a point in changing it, there's no technical benefit. Anything that goes wireless is going to need batteries, fair enough, you have me there, improvements in latency and databus can happen, but in the cabled variety, really?
Can't the public push back and choose to not buy? What can top a set of Sennheiser HD 206's? Anything above that price range isn't going to get any real, material benefit. Digital on the other hand is going to be subject to more variables and I don't honestly think people should be convinced they're getting any improvement there. Anything analogue that isn't 3.5mm are only buying something that has very limited use by comparison. Please world, make this nonsense stop.
I disagree. I think a 'brew' or 'cuppa' are better.
Fair point.
It'd be nice if it could fall back to a last known good config like on Windows (not that that ever works, but the way Ubuntu seems to keep old ones seems like it could be made to work).
Exactly. The headline is rubbish. Part of the kernel installation is to leave the last one in the menu. It's just the new one is a default. If grub is configured with 'savedefault', then the last picked kernel will be chosen for future boots.
What you're saying in the extreme is that things will be free because there are zero costs, but that's not true, you'll still have to pay rates, taxes and materials. Problem is, most of the country will be unemployed so you'll not have a customer, even if things are free for them as they're too busy trying to keep a roof over their heads because, like you, they'll have no income. Zero customers, zero income.