Feel free to use what people give you, but if you want to start making demands, then you need to start making some contributions
The key here is in the word "demand". Asking for features is a contribution in and of itself, but demanding features and complaining if you don't get them, without putting energy into making them happen, is just being a brat.
You should probably upgrade to a more recent version. Redhat 8 is well outdated (I'm not sure Redhat even offers support for it any more, or whether they would if they still offered the Red Hat Linux product line). Fedora and CentOS both maintain Redhat-derived linux distributions that are up to date.
<letter>::= any one of the 52 alphabetic characters A through Z in upper case and a through z in lower case
<digit>::= any one of the ten digits 0 through 9
So a hostname beginning with a digit is actually invalid by the RFC. Nobody actually uses that definition; it turns out it's fairly easy to distinguish a DNS name 127.0.0.com from an IP string 127.0.0.1 in other ways.
If it were Gator, it wouldn't be disabled by default.
Users bounce on [Next >] until they get to the end of the install wizard. People who stop and read the screens on the way are people who stop and read the screens and therefore have no excuse for not knowing what they enabled.
The default configuration could be a bit better, though it is improving. It still has a very conservative memory usage default.
Now that autovacuum is part of postmaster (as of 8.1) it really is almost zero administration for simple databases.
The psql command line client could copy some features from the mysql client:
DESCRIBE and SHOW foo are more intuitive and consistent than \d commands. This can be implemented in the psql client; it's not an SQL query and shouldn't be. Similar for QUIT or EXIT.
It works well in Prime because the developers put some serious attention into the HUD. The displays and UI elements are fixed relative to the helmet, rather than the camera, so they move a bit when you fall or get knocked around. The helmet glass shows your reflection if you're looking towards a bright light.
I don't think anyone's proposing that we distribute from the publicly-editable version, but for a project with sufficient momentum the community behind it may well be able to reach consensus on what snapshots to release for testing. Test snapshots may well have more stringent requirements (eg. all submissions must be vetted, all submissions must pass automated testing, etc) for submissions before being marked "stable" and released.
The basic idea is surprisingly sound, I think, and is worth testing out on a large enough project that it could attract developers. It's certainly not appropriate for every project; the obvious case (small projects) simply won't attract developers and some larger projects have specific concerns that make public access problematic (imagine giving the world write access to, say, OpenSSH).
It's a business problem, not a technical one. To use your example, if both L3 and Cogent peer with QWest, but neither of them has a transit agreement, QWest is under no obligation to move packets from L3 to Cogent or Cogent to L3. It's a case of minimum contractual conformance dictating the technical details.
// note re-declaration for (int i = 0; i < bar; ++i) { ... }
In C++, at least, the scope of entities declared in the for loop is limited to the loop body. Of course, you should avoid using the same loop index twice anyways, in favour of using a more meaningful identifier.
If IPV6 was more widely deployed, you wouldnt have any address problems since IPV6 provides so many addersses that even a home user could have an IPV6 block where the upper 120 bits were fixed and then they would get 8 bits of address to allocate to devices (IANA IPV6 guru so 8 bits for a normal home user might be too much but even 6 bits would give them 64 or so addresses to use)
Actually, you've underestimated by a large margin. The intent is to have the upper 48 bits dictated by the ISP, with the remainder (80 bits) being user-assignable.
Whether anything remotely like this happens in real life remains to be seen, obviously. Consumer ISPs are in love with the one-address-per-customer model, and commercial ISPs tend to charge per address or block thereof.
Do you have any idea what the routing headaches to make that happen would be like? At all? I can't imagine ISPs and backbone providers simply rolling over and soaking the cost, which in turn makes such a bill more expensive to push through and less popular with politicians.
Legitimate edge case: what if you need to have the two domains served by physically separate servers?
Routing devices trying to make decisions based on protocol content (Host: header, for instance) require a huge amount of processing power relative to those that work on the prefix part of the IP address. Good luck telling every business that they need to spend $BIGNUM upgrading the routers they bought, which already cost $BIGNUM in the first place.
Little tangent, but there's a brilliant diary by the guy who wrote Gribbly's Day Out and Paradroid about the development of the latter over on Zzap!64. Fascinating read.
Actually, the machine semi-enslavement and subsequent revolt were Herbert's rationalisation for mentats, humans trained to extreme commputational and logical capacities. His rationalisation for the return of melee combat was the development of the shield, which stopped the vast majority of projectile weapons and destroyed both the weapon and the shield when hit with a laser weapon. That created a situation where the only tenable way to attack someone was extremely impersonally -- nuclear weapons, which he dealt with separately -- or extremely personally: knife and sword combat, where the combatants can time their attacks to slip through a shield, rather than trying to ram through it as a bullet would.
Slashdot Mirror
There's a difference?
The key here is in the word "demand". Asking for features is a contribution in and of itself, but demanding features and complaining if you don't get them, without putting energy into making them happen, is just being a brat.
So... does this explain why Slashdot's Light Mode now looks even shittier?
Xah Lee, is that you?
You should probably upgrade to a more recent version. Redhat 8 is well outdated (I'm not sure Redhat even offers support for it any more, or whether they would if they still offered the Red Hat Linux product line). Fedora and CentOS both maintain Redhat-derived linux distributions that are up to date.
That was Belkin.
Actually, the full grammar in the RFC is
So a hostname beginning with a digit is actually invalid by the RFC. Nobody actually uses that definition; it turns out it's fairly easy to distinguish a DNS name 127.0.0.com from an IP string 127.0.0.1 in other ways.
If it were Gator, it wouldn't be disabled by default.
Users bounce on [Next >] until they get to the end of the install wizard. People who stop and read the screens on the way are people who stop and read the screens and therefore have no excuse for not knowing what they enabled.
I'm not seeing the problem here.
At which point you'd discover something like this.
The default configuration could be a bit better, though it is improving. It still has a very conservative memory usage default.
Now that autovacuum is part of postmaster (as of 8.1) it really is almost zero administration for simple databases.
The psql command line client could copy some features from the mysql client:
DESCRIBE and SHOW foo are more intuitive and consistent than \d commands. This can be implemented in the psql client; it's not an SQL query and shouldn't be. Similar for QUIT or EXIT.
It works well in Prime because the developers put some serious attention into the HUD. The displays and UI elements are fixed relative to the helmet, rather than the camera, so they move a bit when you fall or get knocked around. The helmet glass shows your reflection if you're looking towards a bright light.
Chutzpah.
I don't think anyone's proposing that we distribute from the publicly-editable version, but for a project with sufficient momentum the community behind it may well be able to reach consensus on what snapshots to release for testing. Test snapshots may well have more stringent requirements (eg. all submissions must be vetted, all submissions must pass automated testing, etc) for submissions before being marked "stable" and released.
The basic idea is surprisingly sound, I think, and is worth testing out on a large enough project that it could attract developers. It's certainly not appropriate for every project; the obvious case (small projects) simply won't attract developers and some larger projects have specific concerns that make public access problematic (imagine giving the world write access to, say, OpenSSH).
It's a business problem, not a technical one. To use your example, if both L3 and Cogent peer with QWest, but neither of them has a transit agreement, QWest is under no obligation to move packets from L3 to Cogent or Cogent to L3. It's a case of minimum contractual conformance dictating the technical details.
That would come under the heading of "academic honesty," which is (in theory) one of the things school and university are supposed to instill.
In C++, at least, the scope of entities declared in the for loop is limited to the loop body. Of course, you should avoid using the same loop index twice anyways, in favour of using a more meaningful identifier.
Whoa. I read the article and saw Zonk and something about the history of videogames and assumed it was a dupe.
Your moving navigation bar is really, really distracting.
href="javascript:..." is almost universally a bad idea.
Other than that, that's a really neat idea. How well does it play with JS-disabled browsers?
How would you know?
Actually, you've underestimated by a large margin. The intent is to have the upper 48 bits dictated by the ISP, with the remainder (80 bits) being user-assignable.
Whether anything remotely like this happens in real life remains to be seen, obviously. Consumer ISPs are in love with the one-address-per-customer model, and commercial ISPs tend to charge per address or block thereof.
I'd love to have a rational discussion with you over this, but I feel it's a lost cause. So here's a boggled emoticon:
o_O
Do you have any idea what the routing headaches to make that happen would be like? At all? I can't imagine ISPs and backbone providers simply rolling over and soaking the cost, which in turn makes such a bill more expensive to push through and less popular with politicians.
Legitimate edge case: what if you need to have the two domains served by physically separate servers? Routing devices trying to make decisions based on protocol content (Host: header, for instance) require a huge amount of processing power relative to those that work on the prefix part of the IP address. Good luck telling every business that they need to spend $BIGNUM upgrading the routers they bought, which already cost $BIGNUM in the first place.
Little tangent, but there's a brilliant diary by the guy who wrote Gribbly's Day Out and Paradroid about the development of the latter over on Zzap!64. Fascinating read.
Actually, the machine semi-enslavement and subsequent revolt were Herbert's rationalisation for mentats, humans trained to extreme commputational and logical capacities. His rationalisation for the return of melee combat was the development of the shield, which stopped the vast majority of projectile weapons and destroyed both the weapon and the shield when hit with a laser weapon. That created a situation where the only tenable way to attack someone was extremely impersonally -- nuclear weapons, which he dealt with separately -- or extremely personally: knife and sword combat, where the combatants can time their attacks to slip through a shield, rather than trying to ram through it as a bullet would.