It seems that's for git only, if you want a similar code review product (that's web based) you could look to VMWare's OSS ReviewBoard which is more source-control tool neutral (requires python) and can be automated according to your SCM tool.
They don't need to personalise your search when they know what you're searching for - and they do that based on your search keywords. Personalisation is about amending that list based on what you've searched for previously, you disable it by turning off Web History. This is what they're intending to do.
They're not spying on you for some nefarious purpose, it's to give you better results. You'd probably be a much happier person if you just dealt with it.
Yes, but if they do monitor all web surfing and searches and use the results to target adverts, they'll only be serving ads for porn from now on. How is that going to help society?
Seriously for a moment, once you got ads targetted by the site the ads were displayed on, so if I visited 'lawnmowers.com', I'd want to get ads for lawnmowers and garden supplies. I wouldn't want to get ads targetted things I've been surfing for previously (televisions actually) because I've moved on from that to wanting something new - ie. I wouldn't be getting the ads for stuff I want to buy, only those I had already bought.
The Iranian people deserve better than the lying cheating sack of shits that run the government... at least that's what my Dad told me.
the thing is, his dad told him the exact same thing. Only his generation overthrew the corrupt regime and replaced it with a new one that would bring an era of peace and happiness... oh wait. Maybe 3rd time lucky?
They oversold the bandwidth and they should suffer for it
I agree.
I disagree. overselling is fine, the problem comes when they squeeze too much overselling out of what they've got.
For example, ISP A had 100gb of bandwidth and 1000 customers. They sell each customer 0.1gb, everyone's fine but no customer will use that much bandwidth so most of the network cap is wasted, and when the upstream ISP sells it to you a quite a large sum, you'll find you have no customers as the price you have to charge them is prohibitive.
So, you oversell a bit, and you have 2000 customers, each one getting 0.05gb of bandwidth, most of the time no-one will ever use more that, but occasionally 1 will, when they do a big download. Still, you've made the network more price-competitive but its still probably more than most people will pay for.
So, you go further, and you have 10,000 customers so each could use 0.01gb, which is fine if all they do is surfing and email with the odd bit of streaming and downloading, and the price per customer is great. But then someone decides to up/download 365/24/7... and the model of shared usage falls apart.
The trouble is that most ISPs can look at how much bandwidth they have per customer, and how much actually gets used and make appropriate allowances for that, but most customers will use a lot of bandwidth during 4pm to midnight, and practically none the rest of the day. The better ISPs will tell you this, and will cap you only during that peak time, but the 24/7 downloaders don't care about the other users, so the ISP gets a bit stroppy with them and kicks them off the network, or caps their speeds to persuade them to go or to minimise their impact of the network.
OK, some ISPs oversell to the point where a minimal number of users downloading make an impact, but generally downloads aren't an issue during off-peak times.
The answer is nearly always simple - go with a provider that allows unlimited transfer during off-peak and don't be selfish during the peak times. Its like (car analogy time!) rush hour, drive as fast as you like at 10am, but don't expect to go faster than 5mph at 8am.
The patent troll will probably be set up so that if it loses, it loses the particular patent (which will have no value anymore anyway), pay out what little cash it has, declare itself bust and close. MS would still be owed gazillions of dollars but would not be able to collect any of it.
Patent trollster opens a new company, buys a patent, repeats process.
The only way to beat them would be to make them lose everytime, but then that would require patent laws to be changed so the stupid, obvious patents the trolls use are not valid. But that would cut both ways and a lot of crappy patents the big corporates hold would similarly be invalid.
You can't discriminate against the troll companies, even if they don't make anything from their patents doesn't mean there are other companies who do - and if you have draconian outcomes from a failed patent bid, companies like i4i (who successfully sued Microsoft for blatantly violating their XML searching technology) would probably not feel safe to go to court in the first place.
The only good outcome for geeks is for the patent laws to be revised to something that is fair and sensible. Whilst companies (on both sides of the lawsuits) make money from them, that isn't going to happen.
Active Directory is a MS-standard LDAP server. Like OpenLDAP. You can use LDAP to store your users and login to your linux machines using it as the password store.
However, the OP doesn't understand what AD is, as he thinks it pushes software to clients amongst other things (network configuration?). To get those features you need to buy more server software from MS, like WSUS or Systems Centre. There are similar things available for Linux too (eg a repository for updates, Puppet or CFEngine for client configuration and software pushing).
Perhaps you should have read TFA. The CommonJS spec is the solution to all (well, maybe) the problems you've highlighted.
for example:
Modules are a big deal. CommonJS modules represent the first time a JavaScript community has converged on a module system that is as good as (if not better than) the best of Python and Ruby's--supporting not only dependency management, but also scope isolation and relative module identifiers.
Maybe the spec will improve to remove some of the old Netscape-era incompatibilities, or the browsers will update themselves (like with CSS and HTML) to support only the modern, 'fixed' version. That'd be a good thing.
You'll start using some of the same routines in both client and server.
No, this is the point, you can reuse libraries and similar. With different languages you cannot do this, so have to implement the same thing twice. The biggest problem is that you have to learn 2 languages, you'll be an expert in 1 much sooner than if you have to become an expert in 2 (and to my mind, you never really become a true expert in the second)
and I actually considered using it as a replacement for PHP on the server side in order to keep langauages consistent.
I always thought that would be a good thing - consistency in language means you don't have to learn 5 or so to do meaningful work. Unfortunately, whilst MS appears to agree with me (everything in C#), the rest of the world doesn't. So until we get a javascript world takeover, we'll be stuck with PHP, Perl, Python and Java server-side systems. I'd love for one to win, and JS may as well be the one as its used almost exclusively on the client side too.
(BTW, deleting an array element may reduce the number of elements in the array, but doesn't delete the storage allocated to it - that's probably an optimisation, and isn't such a bad thing).
quite probably, but there are 2 factors that are at issue here:
1. even if you think its poor advice, what can you do? Leave a comment saying "OMG don't do this"? Will anyone read your comment in amongst the ones saying "OMG this is teh best, you solved all my problems, you are so great"
2. I doubt bad advice will be as obvious, more likely you'll see configuration options that end up with your server exposed to the world, maybe set up as an anonymous proxy or given write-access to your ftp server. Just think what configuration someone could give you for your samba shares, or your openldap server!
Whilst that's true, you're forgetting the large amount of 'howto configure xyz' blogs, forums and other sites providing information. Many users don't know how or why the steps they're given work, they just know to follow them blindly. As a result, you can get someone to open their system to you if you were malicious.
So whilst its still not as easy to pwn a linux box, it is still very possible. As the number of users ignorant in system administration increases, this is the attack vector that will become more prevalent. This also applies to a lot of sysadmins, there's a *lot* of stuff in Linux systems today, some of it is very convoluted and difficult to understand let alone configure correctly.
uncounted strings - you mean copy-on-write reference counted ones that require quite a lot of thread safety code inside them (or its a disaster). Some things may seem like an optimisation, but aren't.
Same for non-garbage collected memory. At least those apps don't seem to chew up megabytes of RAM unnecessarily. Again, its not necessarily better.
A better car analogy would be automatic transmission. You don't have to learn how to shift gears, but you do lose on fuel efficiency. Once upon a time nobody cared, but that's not so true today.
Visual BASIC took over C++ and Delphi, as it was easier to program in for managers to understand
Amen. I still remember a manager of mine who got himself a copy of VB, threw together a form and then started talking that programming wasn't so difficult, and why was he paying us programmers so much when coding was so simple anyone could do it.
He never did make that simple form do anything useful, of course.
Hi. The Time travel tourist board called, they said your "work in the future" visa was about to expire and could you make your way back to 1978 please.:)
that's because today's stuff contains libraries that include the things the old timers wrote. No-one needs to write another sort algorithm 'cos its done and is part of the API now. That's the difference, its not that the old ways were inefficient (quite the opposite) but the new guys are standing on the old guys shoulders.
RAM is cheap, but you can only stick so much of it in your motherboard that these is a practical limit, there is also a practical limit on 32-bit CPUs too. In addition, even if you have a 64-bit CPU with unlimited slots, and you use up 500Gig of your super-cheap RAM, how long would you expect your app to take to swap back (not forgetting Windows swaps program code to disk all the time), how long will it take just to load all that bloated code and data into RAM in the first place? How long will it take to iterate through pots of data you've stored in a mistaken attempt to 'optimise through RAM bloat'?
the answer is simple - RAM may be cheap, but if you want to treat it as an unlimited resource and use it without care, your app will still be crap, even on a 500 Gig machine.
I suppose we could go back to the old term for the technical computer people - the "Data Processing" department. I'm not sure when it changed to Information Technology.
No, its not broken for anything other than *Microsoft* virtualisation in Windows Server 2008R2, if you use VMware, then you're fine - it doesn't use the dodgy interrupt that Hyper-V does.
The answer is just not to install Windows Server 2008R2 or use Hyper-V.
I'd far rather pay a bunch of peasants in China to uplift from inefficient farming
you won't when the price of food doubles, said peasants can no longer buy enough to feed themselves, meaning they have to work all hours for even less money leading to mass immigration to 'rich' countries where wages are driven down by immigrant peasants working for far less than the standard rates, leading in turn to mass unemployment, bubbles in shoddy real-estate development, and more tax being raised to support benefits.
A nice equilibrium is what we want, not exploitation based on cheap workers.
switch to it if you really want to, but honestly in ISP and hosting type shops Debian is what I see most
To be fair, what I see in most hosting shops is RedHat (well, Centos). In Enterprises I also see RedHat a lot more than Debian, so I tend to recommend that more than any other distro to people looking to choose a Linux Server OS. For desktops.. I still recommend Ubuntu.
However, if he's comfortable with Debian, stick to using it. There's something to be said for running what you know - thats the primary reason we have so many Windows shops still:)
Slashdot Mirror
It seems that's for git only, if you want a similar code review product (that's web based) you could look to VMWare's OSS ReviewBoard which is more source-control tool neutral (requires python) and can be automated according to your SCM tool.
my favourite is: // TODO: quick hack to make it compile. Needs fixing before checkin
They don't need to personalise your search when they know what you're searching for - and they do that based on your search keywords. Personalisation is about amending that list based on what you've searched for previously, you disable it by turning off Web History. This is what they're intending to do.
They're not spying on you for some nefarious purpose, it's to give you better results. You'd probably be a much happier person if you just dealt with it.
Yes, but if they do monitor all web surfing and searches and use the results to target adverts, they'll only be serving ads for porn from now on. How is that going to help society?
Seriously for a moment, once you got ads targetted by the site the ads were displayed on, so if I visited 'lawnmowers.com', I'd want to get ads for lawnmowers and garden supplies. I wouldn't want to get ads targetted things I've been surfing for previously (televisions actually) because I've moved on from that to wanting something new - ie. I wouldn't be getting the ads for stuff I want to buy, only those I had already bought.
sometimes I want to know what other people know, what they are looking at, what is popular or interesting for them.
so do I, but they complained and now I'm not allowed within 100 yards of them :(
The Iranian people deserve better than the lying cheating sack of shits that run the government ... at least that's what my Dad told me.
the thing is, his dad told him the exact same thing. Only his generation overthrew the corrupt regime and replaced it with a new one that would bring an era of peace and happiness... oh wait. Maybe 3rd time lucky?
They oversold the bandwidth and they should suffer for it
I agree.
I disagree. overselling is fine, the problem comes when they squeeze too much overselling out of what they've got.
For example, ISP A had 100gb of bandwidth and 1000 customers. They sell each customer 0.1gb, everyone's fine but no customer will use that much bandwidth so most of the network cap is wasted, and when the upstream ISP sells it to you a quite a large sum, you'll find you have no customers as the price you have to charge them is prohibitive.
So, you oversell a bit, and you have 2000 customers, each one getting 0.05gb of bandwidth, most of the time no-one will ever use more that, but occasionally 1 will, when they do a big download. Still, you've made the network more price-competitive but its still probably more than most people will pay for.
So, you go further, and you have 10,000 customers so each could use 0.01gb, which is fine if all they do is surfing and email with the odd bit of streaming and downloading, and the price per customer is great. But then someone decides to up/download 365/24/7... and the model of shared usage falls apart.
The trouble is that most ISPs can look at how much bandwidth they have per customer, and how much actually gets used and make appropriate allowances for that, but most customers will use a lot of bandwidth during 4pm to midnight, and practically none the rest of the day. The better ISPs will tell you this, and will cap you only during that peak time, but the 24/7 downloaders don't care about the other users, so the ISP gets a bit stroppy with them and kicks them off the network, or caps their speeds to persuade them to go or to minimise their impact of the network.
OK, some ISPs oversell to the point where a minimal number of users downloading make an impact, but generally downloads aren't an issue during off-peak times.
The answer is nearly always simple - go with a provider that allows unlimited transfer during off-peak and don't be selfish during the peak times. Its like (car analogy time!) rush hour, drive as fast as you like at 10am, but don't expect to go faster than 5mph at 8am.
The patent troll will probably be set up so that if it loses, it loses the particular patent (which will have no value anymore anyway), pay out what little cash it has, declare itself bust and close. MS would still be owed gazillions of dollars but would not be able to collect any of it.
Patent trollster opens a new company, buys a patent, repeats process.
The only way to beat them would be to make them lose everytime, but then that would require patent laws to be changed so the stupid, obvious patents the trolls use are not valid. But that would cut both ways and a lot of crappy patents the big corporates hold would similarly be invalid.
You can't discriminate against the troll companies, even if they don't make anything from their patents doesn't mean there are other companies who do - and if you have draconian outcomes from a failed patent bid, companies like i4i (who successfully sued Microsoft for blatantly violating their XML searching technology) would probably not feel safe to go to court in the first place.
The only good outcome for geeks is for the patent laws to be revised to something that is fair and sensible. Whilst companies (on both sides of the lawsuits) make money from them, that isn't going to happen.
such as DRM. Ok, not a feature I'd like to have, but it means a lot of content will not be available on non-Silverlight devices.
Active Directory is a MS-standard LDAP server. Like OpenLDAP. You can use LDAP to store your users and login to your linux machines using it as the password store.
However, the OP doesn't understand what AD is, as he thinks it pushes software to clients amongst other things (network configuration?). To get those features you need to buy more server software from MS, like WSUS or Systems Centre. There are similar things available for Linux too (eg a repository for updates, Puppet or CFEngine for client configuration and software pushing).
Perhaps you should have read TFA. The CommonJS spec is the solution to all (well, maybe) the problems you've highlighted.
for example:
Modules are a big deal. CommonJS modules represent the first time a JavaScript community has converged on a module system that is as good as (if not better than) the best of Python and Ruby's--supporting not only dependency management, but also scope isolation and relative module identifiers.
Maybe the spec will improve to remove some of the old Netscape-era incompatibilities, or the browsers will update themselves (like with CSS and HTML) to support only the modern, 'fixed' version. That'd be a good thing.
You'll start using some of the same routines in both client and server.
No, this is the point, you can reuse libraries and similar. With different languages you cannot do this, so have to implement the same thing twice. The biggest problem is that you have to learn 2 languages, you'll be an expert in 1 much sooner than if you have to become an expert in 2 (and to my mind, you never really become a true expert in the second)
and I actually considered using it as a replacement for PHP on the server side in order to keep langauages consistent.
I always thought that would be a good thing - consistency in language means you don't have to learn 5 or so to do meaningful work. Unfortunately, whilst MS appears to agree with me (everything in C#), the rest of the world doesn't. So until we get a javascript world takeover, we'll be stuck with PHP, Perl, Python and Java server-side systems. I'd love for one to win, and JS may as well be the one as its used almost exclusively on the client side too.
(BTW, deleting an array element may reduce the number of elements in the array, but doesn't delete the storage allocated to it - that's probably an optimisation, and isn't such a bad thing).
quite probably, but there are 2 factors that are at issue here:
1. even if you think its poor advice, what can you do? Leave a comment saying "OMG don't do this"? Will anyone read your comment in amongst the ones saying "OMG this is teh best, you solved all my problems, you are so great"
2. I doubt bad advice will be as obvious, more likely you'll see configuration options that end up with your server exposed to the world, maybe set up as an anonymous proxy or given write-access to your ftp server. Just think what configuration someone could give you for your samba shares, or your openldap server!
Whilst that's true, you're forgetting the large amount of 'howto configure xyz' blogs, forums and other sites providing information. Many users don't know how or why the steps they're given work, they just know to follow them blindly. As a result, you can get someone to open their system to you if you were malicious.
So whilst its still not as easy to pwn a linux box, it is still very possible. As the number of users ignorant in system administration increases, this is the attack vector that will become more prevalent. This also applies to a lot of sysadmins, there's a *lot* of stuff in Linux systems today, some of it is very convoluted and difficult to understand let alone configure correctly.
uncounted strings - you mean copy-on-write reference counted ones that require quite a lot of thread safety code inside them (or its a disaster). Some things may seem like an optimisation, but aren't.
Same for non-garbage collected memory. At least those apps don't seem to chew up megabytes of RAM unnecessarily. Again, its not necessarily better.
A better car analogy would be automatic transmission. You don't have to learn how to shift gears, but you do lose on fuel efficiency. Once upon a time nobody cared, but that's not so true today.
Visual BASIC took over C++ and Delphi, as it was easier to program in for managers to understand
Amen. I still remember a manager of mine who got himself a copy of VB, threw together a form and then started talking that programming wasn't so difficult, and why was he paying us programmers so much when coding was so simple anyone could do it.
He never did make that simple form do anything useful, of course.
you really should try and use 80 columns only
Hi. The Time travel tourist board called, they said your "work in the future" visa was about to expire and could you make your way back to 1978 please. :)
that's because today's stuff contains libraries that include the things the old timers wrote. No-one needs to write another sort algorithm 'cos its done and is part of the API now. That's the difference, its not that the old ways were inefficient (quite the opposite) but the new guys are standing on the old guys shoulders.
RAM is cheap, but you can only stick so much of it in your motherboard that these is a practical limit, there is also a practical limit on 32-bit CPUs too. In addition, even if you have a 64-bit CPU with unlimited slots, and you use up 500Gig of your super-cheap RAM, how long would you expect your app to take to swap back (not forgetting Windows swaps program code to disk all the time), how long will it take just to load all that bloated code and data into RAM in the first place? How long will it take to iterate through pots of data you've stored in a mistaken attempt to 'optimise through RAM bloat'?
the answer is simple - RAM may be cheap, but if you want to treat it as an unlimited resource and use it without care, your app will still be crap, even on a 500 Gig machine.
I suppose we could go back to the old term for the technical computer people - the "Data Processing" department. I'm not sure when it changed to Information Technology.
No, its not broken for anything other than *Microsoft* virtualisation in Windows Server 2008R2, if you use VMware, then you're fine - it doesn't use the dodgy interrupt that Hyper-V does.
The answer is just not to install Windows Server 2008R2 or use Hyper-V.
Really, I wonder who even buys the "trickle down" nonsense anymore.
Rich people. The same ones who are driving this imbalance in trade and work.
I'd far rather pay a bunch of peasants in China to uplift from inefficient farming
you won't when the price of food doubles, said peasants can no longer buy enough to feed themselves, meaning they have to work all hours for even less money leading to mass immigration to 'rich' countries where wages are driven down by immigrant peasants working for far less than the standard rates, leading in turn to mass unemployment, bubbles in shoddy real-estate development, and more tax being raised to support benefits.
A nice equilibrium is what we want, not exploitation based on cheap workers.
switch to it if you really want to, but honestly in ISP and hosting type shops Debian is what I see most
To be fair, what I see in most hosting shops is RedHat (well, Centos). In Enterprises I also see RedHat a lot more than Debian, so I tend to recommend that more than any other distro to people looking to choose a Linux Server OS. For desktops.. I still recommend Ubuntu.
However, if he's comfortable with Debian, stick to using it. There's something to be said for running what you know - thats the primary reason we have so many Windows shops still :)