Slashdot Mirror


User: turbidostato

turbidostato's activity in the archive.

Stories
0
Comments
5,722
First seen
Last seen
Profile
(view on slashdot.org)

Comments · 5,722

  1. Re:Do they or do they not have the source legally? on Zeta Goes Gold · · Score: 1

    "That's an interesting perspective, but how does that increase your freedom?"

    Of course yes. In exactly the same way your consitutional right for free spech does.

    "It would seem to me that for most very large OSS projects with large codebases (>100K loc) I would be able to code my own before I could learn the codebase well enough to make non-trivial changes."

    So what? From time to time I have to make trivial changes to some other's code base, and know what? they maybe are trivial because of the code effort, but to me they are very important because they mean the difference between being able to properly use the app or not, and I can do those changes because it is open source. Why do you thing that liberty comes only when you efectively exercise it, or only when you excercise it to a great extent?

    A recent practical case of me: I have to support and old privative app. It is a little nightmare when I have to upgrade glibc when probably just by recompiling most of the problems would go away, or just making little changes to stay compatible with the API. Know what? I can't do it because I haven't access to the source code, mush less the ability to modify it, no matter if it is for short or for large.

    "The only custom builds of FF I know of are Netscape 8"

    The only custom build of IE I know of is... noone.

    "Linux distros? How many of them make major changes to the kernel?"

    Red Hat for one. Do you know what? Red Hat didn't develop the base of the kernel, Linus Torvalds did. Still Red Hat can and do modify the kernel because it is open source. What company do you know that is for Windows the same than Red Hat is for Linux? Exactly: noone, because only Microsoft has that privilege. And even if you wouldn't ever modify Windows kernel even if it were open, you can be sure you are paying for that in the form of vendor locking and substandard development (since Microsoft is not pressed by competence).

    "How many of them are released by a single person?"

    Slackware for one. And it is only the strongestly stablished; you can find a "ton" of short niche one person (or almost) distributions if you go to distrowatch.

    "Yeah, it may be a huge freedom, but what does it matter"

    Yeah... do you know something? I come from a country which was under a dictatorship until some years ago, and then, you could hear quite a lot of people saying exactly the same you do: what does it matter free speech? After all I am not a reporter. What does it matter union or strike rights, after all I'm quite happy with my job. What does it matter justice being corrupt, after all I'm a lawful unimportant citizen which never has been near to a judge and, by all bets, will never be.

    Do you know what's the answer to those people? You are a poor ignorant.

    Now you know my answer to you.

  2. Re:Do they or do they not have the source legally? on Zeta Goes Gold · · Score: 1

    "Hiring somebody to modify the code can cost far more than the alternative proprietary product."

    Yes.

    Or it can cost you less.

    Anycase, since Apache's source code is free as in free speech YOU CAN CHOOSE. You can't choose over proprietary code.

    So what is your point?

  3. Re:No one cares on Debian GNU/Linux now in AMD64 form · · Score: 2, Insightful

    Or maybe this is no news for anyone that cares about Debian on AMD64.

  4. Re:whatever... on Where is the Killer Calendar? · · Score: 1

    When you are talking about managed IT, central servers are good, information spreaded an unmanaged over a lot of clients is bad. Having groupware servers is good, using them propietary protocols and formats is bad.

    Still, this doesn't mean clients can't cache relevant information so you don't have to go to the server for every little operation, provided it is developed properly. Servers like Kollab can use disconnected IMAP for this, for instance.

    "Wasn't there a ximian exchange client?"

    But you are still using a proprietary server with proprietary patent-encumbered comunication protocols. Noone apart from Microsoft can insure proper functionality (not that Microsoft will insure it, but that's a different issue). Vendor locking is bad.

  5. Re:the real need? on Where is the Killer Calendar? · · Score: 1

    "The real strength of Outlook, (as it has been mentioned before,) is really it's connection to an Exchange server. The problem is that it ties you to an Exchange server."

    It ties you and your whole organization to an Exchange server... and a Windows Server, and Windows clients.

  6. Re:Until its shareable... on Where is the Killer Calendar? · · Score: 1

    "My shop recently switched over to Exchange from a sweet Qmail/vpopmail toaster setup, solely because we wanted the shared calendar"

    As if there is no open source shared calendar or even whole groupware suites (like eGroupware or Kolab) over there.

    "And that's what the Open Source world is seriously missing."

    I'd say it is proper knowledge from your side about suitable open source solutions that is seriously missing.

    "When, oh, when, is someone going to code a decent exchange replacement that does not require over 100 RPM's to install?"

    And what would be the matter with a proper solution even if it requires 100 RPM's to install? You just told Qmail/vpopmail was a decent e-mail solution for you, and I bet the box you had for this purpose did require well over 100 RPM's, and you didn't saw a problem with that!

  7. Re:Who's got nested tasks? on Where is the Killer Calendar? · · Score: 1

    KOrganizer coupled with Karm (from the KDE PIM package) do this (categories, nested tasks, time spent, start and due dates, percentage done, task members, attached files...) while still not perfectly: they still step each other, but it will work fine if you have opened only one of them at a time.

  8. Re:Talk about over-engineering, people on Where is the Killer Calendar? · · Score: 1

    "a small attache journal and a fountain pen"

    Yes, it makes quite an interesting technogadget. But being small and portable makes it easy to lose it.

    When was last time you backuped it just in case?

  9. Re:For OS X: Entourage 2004 on Where is the Killer Calendar? · · Score: 1

    "Having one "Monolithic" application for email/calendar is rather nice to have"

    And yet, having the ability to choose if you prefer having a big monolithic app, or a few bunch of apps you can open and close at leisure is probably better.

    And that's what KDE's PIM package gives to you: you can either use the "monolithic" Kontact "shell" or you can open separatly the e-mail app, the news one, the Calendar, etc.

    "...and Outlook is really the only good program at communicating with an Exchange server"

    And that's probably a very big reason NOT to use Exchange. I thought we all learnt about how evil vendor locking was back in the 70's-80's.

  10. Re:Korganizer on Where is the Killer Calendar? · · Score: 1

    "I use Korganizer synced with a palm for my scheduling"

    Me too, and it's great!

    "Like any piece of software, it's far from perfect"

    And yet, only problems I see its Knode's problems that makes it growing the window beyond the screen limit from time to time, and Karm still not integrating properly with Korganizer so timings and subtasks don't appear within ToDo list or the calendar window (well, you can kind of hack it out, but then you are risking the integrity of karm's ics file).

    I do want to believe those that say Outlook 2003 is such a great app but:
    1/ I can't use it except on a single propietary OS (I don't think vendor locking can help me too much).
    2/ I can't properly use it to its full extent except with a single propietary groupware server (I don't think vendor locking and proprietary data formats can help me too much).
    3/ I don't miss anything too important on my current solution (and it doesn't impose me a single vendor, it uses publicly avaliabe data formats, and it is open source and free as in free speech and as in free beer too).

  11. Re:Outlook 2003 on Where is the Killer Calendar? · · Score: 1

    Well, yes, kindof...

    Let's an amazingly great design: that from Windows GUI. When I try to start Internet Explorer on my Linux system it is quite strimlined... oh, wait...

  12. Re:Outlook 2003 on Where is the Killer Calendar? · · Score: 1

    "If all the developers developing functionality in this KDE-only style would put the (wasted) effort into programming this on the OS-level"

    Which OS? Linux? FreeBSD? NetBSD? OpenBSD? Solaris?

  13. Re:If you want to work it out.. on Realistic Sysadmin Workload for a Company of 30? · · Score: 1

    "Then cut a deal that if after a month, if your time spent per week is more than 3%, he'll hire a professional Admin"

    Plainly stupid. This suggestion is only good to test the case. If ever his boss accepts this deal is a proof beyond all doubt he must run out as fast as possible.

    Think it for a moment: "If I spend more than 3% on sysadmin this month you'll hire a professional Admin". Can it make ANY meaning? After a month you go and show:

    -Hey, I told you: see, I spent not only 3% but 10% on sysadmin*1, so you must hire a professional one
    -OK, lad: so you spent 10% of your time as sysadmin and you think I'm going to hire a professional one, which obviously will do his work faster than you, knowing there's only work for about 5% his time? Now tell me how it makes economical sense paying 100% of a techie for only 5% of his workforce.
    -But, but... you promised me...
    -But, but... return to work you idiot, continue with your sysadmin tasks and remember you are late on your current developing project!

    *1 As this is just the first month, people still didn't know you were the new "gafferboy" and they let you quite easy; you were lucky that nothing broke and, of course, you didn'd attack anyone of the 300 "little things" like proper backups, that were awaiting since Reagan's presidency days -after all you were "resigning" from sysadmin in a month, so that explains why it only took you 10% of your time.

  14. Re:If you take it, say goodbye to programming on Realistic Sysadmin Workload for a Company of 30? · · Score: 1

    "Assumption: The company has plans to grow at the new location (else why did they move?)"

    They bougth the company because their main building is downtown; they sold it for a big bunch of money, then they are moving to the suburbs to a cheaper near-to-nothing facilities about half the size, since for now on they are outsourcing to China.

    As the sysadmin load diminshes due to outsourcing, one of the first you be fired will be the sysadmin, since, after all, it's only 1% of a typical developer time, and diminishing since there's less and less work here. On the other hand, bad results on the central will be the chance for further outsourcing, thus even less sysadmin work.

    In about one year and a half, when the new bosses have squeezed as much money they could from dismantling the company and recovering their stocks, they'll just fire the last two developers and close the circus.

    Feel better now you know?

  15. Re:Depends if you do the optional extras on Realistic Sysadmin Workload for a Company of 30? · · Score: 1

    Everything quite true.

    It is for a reason that I LOVE this job!!!

    (/me ducks)

  16. Re:hiring? on Realistic Sysadmin Workload for a Company of 30? · · Score: 1

    "Are they hiring, I'm qualified for that job!"

    Are you sure you want the post? Remember they want to pay 1% of usual wages!

  17. Re:May be possible... on Realistic Sysadmin Workload for a Company of 30? · · Score: 1

    "I work for a (small) company that manages the whole IT infrastructure for several other small companies [...] only require approx 1 hr of our time after it is set up"

    But we musn't forget that:
    a) You are experts that do that for a living
    b) Properly managed sysadmin scales quite a lot (so properly managed 100 clients don't take much more time than 30... on the other hand, 30 clients don't take much less that 100, if you see my point)
    c) This guy has to account for setting up times too
    d) Very probably his environment (a programming shop) is not very suitable for the kind of solutions you use on service/office companies

    How many man-hours of experience does your staff accumulates that the guy of the article lacks of?

    How many hours did it take to your company to "do it right" so now you have that low maintenance time?

    How many hours are "shared" among more than one client that would have to be expended even if you only had one client? (monitoring/mirroring/automation tools, shared servers etc.)?

    Now, just as an example: some other people told he would use a helpdesk kind of tool so he can track incidents and (maybe) avoid people addressing directly to him (breaking his programming concentration). He is not an expert; even if the box is already running (OS, web server, name services, etc.) deploying something like RT, phpHelpDesk or something like that can very well take him four hours. But four hours is his allotted time for about two months, and he hasn't even started his duties!

    Now, I recently installed four servers for a client, migrating/upgrading dependant services, etc. Just documenting the installation made up six hours (about 8% of total time), so there go other three months of allotted time.

  18. Re:Insane on Realistic Sysadmin Workload for a Company of 30? · · Score: 1

    Well, that's the exact opposite of the article.

    And then I'd say that although may be your special environment really needs such an overloaded IT staff, something smells uglily wrong there. One IT guy each 20 people? Sounds quite excesive under usual circumnstances.

  19. Re:BAD ADVICE on Realistic Sysadmin Workload for a Company of 30? · · Score: 1

    "Just be prepared for the unemployment line"

    I don't think this is the case here. Of course his rejection won't make his boss too happy, but he is a programer among 30; unless he has been offered the task because he is considered to be the less productive of the develop team (in which case he will be fired soon no matter what), the task will be offered to anyone else (perhaps he can try the "I heard [mate] has a home network in his basement" trick) or an outside consultor will be contracted. In a month of two the incident will be forgotten.

    "And why would you give up the opportunity..."

    While it seems reasonable under general condition, under given circumnstances it seems there's much more to loose than to win by accepting that new task.

  20. Re:IT is to laugh... or cry. on Realistic Sysadmin Workload for a Company of 30? · · Score: 1

    "It's not clear whether you're expected to be the systems administrator, the network engineer, or the all-purpose all-singing all-dancing IT guy"

    On an about 30-50 people shop, it is out of question: you are the Mr half-a-dozen-hats there.

    "Tell your boss..."

    The problem is what can you tell to such a kind of a boss. Probably the easiest thing would be in the lines of "OK, I'll take your numbers: I'll start next Monday". Then on Monday, just start your usual programming day; as soon as you are called-on-duty as a sysadmin, you start your crono. As soon as it marks 24 minutes of "sysadmining" (the alotted pack for the week) you just ring your bell and say "Here is the Fires-In-Progress paper: everybody just needs to write down their sysadmin-related problems here and I'll start with them punctually and in due order next Monday at 9AM".

    On a Happy World it wouldn't take to much for the boss to see kind of a trend and reconsider his opinions regarding sysadmin efforts and taking you apart to tell you how right you were.

    On our Really Wrecked World, you would be fired by Friday (and this is true for almost all of the "what you should do..." messages here, except those saying "don't take it").

  21. Re:Do not accept on Realistic Sysadmin Workload for a Company of 30? · · Score: 2, Interesting

    "Make a list of activities he wants done and activities that are done weekly in a normal sysadmin job"

    If his boss is the kind of boss that says supporting 30 developer computers takes 24min a week, that strategy won't do the trick. I can see it:
    -So, let's see what do you want the sysadmin to do
    -Humm... a big bunch of nothing, things are well enough the way they are, so he won't have to touch anything. In fact, the less he touches, the less he can break apart.
    -So nothing by zero sums up to... zero minutes/week
    -See!!!??? And then I'm so generous to give to you 24min/week just in case!

    "As someone else said, a sysadmin does not directly show up in a profit statement"

    The problem is even worse: sysadmin job, specially if it is well done does not directly show up -full stop. It is akind to an iceberg; 90% is hidden for untrained eyes as are those both of the boss and our developer. And then, just by saying it will take no more than 24min/week that boss has already shown how irrespective he is about that kind of job, so you much better stay away from it: nothing good can come out for you!

    "Also, once you get started on sysadmin work, start logging your time."

    That's a terribly difficult task, probably impossible under current environment. For you to be able to properly defend your logged times (properly by that boss' opinion) you can only spend time suffocating fires (you won't be able to tell him "I was reading our router's admin manual just in case" since that boss will reply "is it broken? then you don't need to waste your time that way"), but as soon as you go that dynamics it is impossible you can do your work properly and both your boss and your mates will be really upset with you (since you will be a) the guy that allows for things to break; b) the guy that always comes late -of course, you are working on other fires; and c) the guy that takes an eternity to fix everything -how not! since you are unproperly trained, time perception from your "victims" is heavily accelerated since the work they can't do was to be finished yersterday, no time is alotted for you to train yourself and you are called when the thing is really messed up and everybody is shouting at you and looking over your shoulder).

    There's a reason why sysadmin is a profession: it is both technically and psicologycally challenging, and it is a reason there are organizations like the SAGE: it is difficult for untrained people to see that sysadmins do really deserve their wages.

  22. Re:Do not accept on Realistic Sysadmin Workload for a Company of 30? · · Score: 1

    "accept it - but make clear that you won't do unpaid overtime to meet requirements of both positions. the employer is likely paying you for 8 hours a day, so give him that."

    But then, as wisely has been pointed out, neither his sysadmin tasks nor his programming tasks will go up-to-date, probably not even if his 40hours/week become 70hours/week, and by accepting the task he more than probably will be considered to be an under-average employee.

    How is it that on a technical company the boss says sysadmin will be 1% of a man/week, that is 40/100=24minutes/week!!!???

    Your boss is either dumb, and you most probably will be fired in a 6 to 12 months span, since you will be percieved as a very subpar professional, or he is "too clever" and want to press you to burnout. Either posibility is not to be interesting for you.

    The wisest choice already has been highlighted: DON'T TAKE IT!!! You can tell your boss that surely you are not-not-not up to the task, and that at the very least a professional-grade sysadmin should be contracted to look after the site migration and tying things properly in order to minimize maintenance load afterwards. If terribly pressed, promise you boss that ONCE THE SITUATION IS STABILIZED, NOT A SECOND EARLIER, you will go after the sysadmin, and learn how to maintain the installation from him.

    Notwithstanding your boss' opinion, this will take no less than four to six months; if your boss is either "too dumb" or "too clever" as per the previous paragraph, you will see clearly on the contracted sysadmin's flesh (is he cared properly? is his job properly valued) and you will learn not to accept the job at any rate; if your boss was honest, just didn't know how hard proper sysadmin is, he will have learned by now, and then you and your boss will be able to reconsider the situation under proper lights.

    As an extra point, there's only one way to achieve your boss' dreamed 1%man load and it is by really ironing boxes configuration and changing nothing ever, and even then, with 30 boxes just hardware-related problems migth go well over 24min/week on average. But then, those 30 boxes are mainly developer boxes which usually can't be "tied up" so hard for the very needs of the developers, so it is obvious that even if developers manage to "autosupport" most of the time (on a low effectiveness manner, since they don't know so well that kind of job -aren't they payed for programming, why are they loosing their precious time installing security patches, reinstalling drivers, etc. on their boxes?), when they wreak havoc, they do the great way, with fireworks too.

    I'll tell it again: unless you want to change careers and go into sysadmin, don't take that pill, and even if you want to change careers, your boss doesn't seem to be the proper one to give it a try.

  23. Re:I don't understand... on Microsoft's Slap at Samba · · Score: 2, Interesting

    "If MS had 50% of the market, SAMBA wouldn't have a case"

    Even in that case Samba would have a winning case.

    You have ALWAYS to allow for interoperatibility; if you don't like it, just don't play the game. Bill Gates is free to take his money and smoke it if he wants (well, even literally, I think that in the States firing a money note is not considered "destroying federal property" now), but if he wants to "build thingies" that interoperaty with anything else, he should allow for anything else to interoperate with hims.

    Even in the USA with such corporate-favouring laws you should have present the "Lexmark case". I don't think Lexmark holds 50% of any bussiness, still they lost regarding the "unofficial" tonner marketeers.

  24. Re:Things Man Was Not Meant To Know on Study Links Genetic Diseases to Intelligence · · Score: 2, Insightful

    "Ignorance is natural too, but I don't see anyone arguing that people who've learned nothing should be considered equal to those who have studied extensively and developed important intellectual skills."

    "One man, one vote" anyone, uh?

  25. Re:The top is not an issue on Study Links Genetic Diseases to Intelligence · · Score: 1

    "But you can still say that Group A is mucher better on average that Group B, so much so that a few standard deviations of Group A are completely above a few standard deviations of Group B."

    And you think that if such were the case you'd need meaures from a highly knowledgeable bunch of scientists to realize it.

    "This is about discovering something about ourselves that could affect society in a fundametnal way"

    What you seem to forget is that highly relevant things tend to be self-evident.

    The fact that your simple untrained eye is enough to see baboons are less inteligent that (most) people, makes obvious that [put your favorite ethnic group here] can't be soooo much "inteligent" (whatever the exact definition is) than your [hathred ethinc group here] or else you wouldn't need a group of scientists to tell you.