I too, was not impressed with Seifreid's followup. He changed his arguments on a number of different issues, and completely avoided areas where he was shown to be wrong or misleading.
Agreed.
However; I think he has a point regarding the missing facilities in the SSH protocol for expiring of keys. Keys should not live forever. And if you can't distinguish a changed key (a potentional security breach) from an expired key (a normal event in a security conscious environment), you and your users got a problem.
At my compagny we are use a two-version policy: By default we're using the latest production quality version (6.2) but for development/experimentation we're using the latest and greatest (7.0).
We won't normally ship a RH 7.0 server (or x.0 at all), but when we start a new project we're always using the latest version. Given some 6 to 12 months development time the once immature version would have grown up and reached the production quality level.
We're currently working on a project that started out on a 6.0 but now runs on a 6.2. The 6-months release cycle of RedHat fits our business very well.
Oh, btw: There is an exception to the never-ship-a-x.0 rule - today I completed a performance tuning of a Oracle database. Previously it ran on NT. Now it runs on RH 7.0 with astonishing results and our customer demands that we ship him the RH 7.0 version:-)
I have not read The Discourses but I rate The Prince on the top-ten of the best books I've ever read.
For those who have not read it, I would describe it as an empiric investigation of the human mind. It is so refreshingly free of all the usual pretty-painting of the world and the mechanisms that keeps it together. Instead it focusses on how people are instead of how they ought to be.
>you go from being a good technician to lousy manager with the click of fingers
I switched from tech to management just a half year ago, and thats exactly how I
feel.
There is a big difference between being a good tech and being a good manager. A good tech is solving problems and is rewarded accordingly. A good manager is dealing with problems; moving them around, pushing them onto anothers disk, hiding them for the customers etc. etc. The success criteria seems a lot
more blurry but if you're doing resonable and avoids the worst disasters you will be given time. Management is hard - good management is even harder - and it takes time to learn.
Another big difference is how you manage truth: As a techie you are expected to tell the truth. If I was to cooperate with you and I downright lied about the interface specification between our modules, I would be a lousy programmer. OTOH if I was telling our customers the truth about the status of our projects I would be
a lousy manager. It takes some time to get accustomized to:-)
Not that I didn't wanted to enter management - oh no - but I was taken somewhat by surprise by the new requirements.
BTW: Those 20 percent you were talking about ? Forget it ! You should expect at least 50 percent and most likely 80 percent until you are a little more experienced.
>Is this just a new version of gcc, or something else more noteworthy?
Its "just" a new version of gcc. Or rather: Its a special RedHat version of the not-so-soon to be released GCC 3.0 compiler. GCC 3.0 have been worked on for a while now, but the official release is still some months away.
RedHat is probably using some of their in-house expertice (the Cygnus folks) to rip off the most stable parts of the gcc snapshot and package it for their new distribution.
To my knowledge there is nothing revolutionary about the GCC 3.0 - except for better standard compliance and perhaps better optimizations.
I have come to hate these talks with my boss every half a year. Each and every time RedHat is releasing a new version I am going to disturb him with (what he thinks) is irrelevant details.
He is not exactly entusiatic about these linux machines I keep running here and there. He would certainly prefer to run Windows like everyone else, but as I have told him: They're cheap and they're doing the job - flawlessly. So they stay. He even tries to ignore that once a while a new one pops up somewhere on our network.
But those half-yearly "boss-I-need-a-few-bucks" talks is humilitating for me and just plain irritating for him.
So this subscription model will require a once-and-for-all decision about running linux or not, and then our financial department will take care of the rest. That's the way it should be !
Slashdot Mirror
Agreed.
However; I think he has a point regarding the missing facilities in the SSH protocol for expiring of keys. Keys should not live forever. And if you can't distinguish a changed key (a potentional security breach) from an expired key (a normal event in a security conscious environment), you and your users got a problem.
We won't normally ship a RH 7.0 server (or x.0 at all), but when we start a new project we're always using the latest version. Given some 6 to 12 months development time the once immature version would have grown up and reached the production quality level.
We're currently working on a project that started out on a 6.0 but now runs on a 6.2. The 6-months release cycle of RedHat fits our business very well.
Oh, btw: There is an exception to the never-ship-a-x.0 rule - today I completed a performance tuning of a Oracle database. Previously it ran on NT. Now it runs on RH 7.0 with astonishing results and our customer demands that we ship him the RH 7.0 version :-)
For those who have not read it, I would describe it as an empiric investigation of the human mind. It is so refreshingly free of all the usual pretty-painting of the world and the mechanisms that keeps it together. Instead it focusses on how people are instead of how they ought to be.
Truely recommendable
I switched from tech to management just a half year ago, and thats exactly how I feel.
There is a big difference between being a good tech and being a good manager. A good tech is solving problems and is rewarded accordingly. A good manager is dealing with problems; moving them around, pushing them onto anothers disk, hiding them for the customers etc. etc. The success criteria seems a lot more blurry but if you're doing resonable and avoids the worst disasters you will be given time. Management is hard - good management is even harder - and it takes time to learn.
Another big difference is how you manage truth: As a techie you are expected to tell the truth. If I was to cooperate with you and I downright lied about the interface specification between our modules, I would be a lousy programmer. OTOH if I was telling our customers the truth about the status of our projects I would be a lousy manager. It takes some time to get accustomized to :-)
Not that I didn't wanted to enter management - oh no - but I was taken somewhat by surprise by the new requirements.
BTW: Those 20 percent you were talking about ? Forget it ! You should expect at least 50 percent and most likely 80 percent until you are a little more experienced.
Anyway: Go for it ! And good luck.
Its "just" a new version of gcc. Or rather: Its a special RedHat version of the not-so-soon to be released GCC 3.0 compiler. GCC 3.0 have been worked on for a while now, but the official release is still some months away.
RedHat is probably using some of their in-house expertice (the Cygnus folks) to rip off the most stable parts of the gcc snapshot and package it for their new distribution.
To my knowledge there is nothing revolutionary about the GCC 3.0 - except for better standard compliance and perhaps better optimizations.
Small business does too.
I have come to hate these talks with my boss every half a year. Each and every time RedHat is releasing a new version I am going to disturb him with (what he thinks) is irrelevant details.
He is not exactly entusiatic about these linux machines I keep running here and there. He would certainly prefer to run Windows like everyone else, but as I have told him: They're cheap and they're doing the job - flawlessly. So they stay. He even tries to ignore that once a while a new one pops up somewhere on our network. But those half-yearly "boss-I-need-a-few-bucks" talks is humilitating for me and just plain irritating for him.
So this subscription model will require a once-and-for-all decision about running linux or not, and then our financial department will take care of the rest. That's the way it should be !