Open Source the new name for crappy software



  • I needed a windows FTP server just to quickly lash up a test - a quick google got me downloading Filezilla.

    I suppore the first suprise is that it didn't help me create a rule in windows firewall.  No problem, sorted that out.

    FTP Server is installed as a service, however every time i start the PC it asks to connect to the server admin console.  I had a quick look for an option to turn that off and I couldn't find one so a had a quick google to find the answer.  Only to find others asking the same question for the last 6 years and the following answer:

     



  • You think that's crappy OpenSource? Seriously, try using the StupidFTPD server. You'll get security problems and crashes for free! (Yes, that's a real open source project)



  • @Helix said:

    I needed a windows FTP server just to quickly lash up a test - a quick google got me downloading Filezilla.

    I suppore the first suprise is that it didn't help me create a rule in windows firewall.  No problem, sorted that out.

    FTP Server is installed as a service, however every time i start the PC it asks to connect to the server admin console.  I had a quick look for an option to turn that off and I couldn't find one so a had a quick google to find the answer.  Only to find others asking the same question for the last 6 years and the following answer:

     

    http://trac.filezilla-project.org/ticket/2662

    Welcome to the world of open-source software, where a feature the creators of the software don't care about gets ignored, regardless of how useful or necessary said feature would be for users.



  •  Well, I'll admit that I don't like software that does this, but I've come across a number of packages that are only configured using their installers. It's a valid chioce used by a lot of comercial software as well.

     If you don't like it: well it's open source, so you can change it however you like, can't you?



  • @Helix said:

    I suppore the first suprise is that it didn't help me create a rule in windows firewall.
     

    Why is that surprising? I can't see how one of the requirements for FTP server software is "assist the user in operating the firewall of their choice".
    However, you've encountered a common problem in many open-source projects:

    • developers concentrating on adding functionality for their own usage
    • developers focussing upon coding and neglecting documentation (this isn't just an open-source problem, but many open-source projects don't have serious management and stringent policies so there's less control over quality)
    • volunteers drifting off-project, leaving things incomplete.
    All I can say is: don't judge all open-source by your experiences with this one utility. Open-source simply means the source code is available - it doesn't necessarily imply free or volunteer-driven projects; those are just characteristics associated with open-source S/W.



  • @robbak said:

     Well, I'll admit that I don't like software that does this, but I've come across a number of packages that are only configured using their installers. It's a valid chioce used by a lot of comercial software as well.

     If you don't like it: well it's open source, so you can change it however you like, can't you?

    Of course you can. And then you submit a patch for the change and the maintainer rejects it because they can't be arsed to merge it in, or you made a 1-space mistake that breaks the project's coding conventions and the maintainer can't be arsed to fix it himself, or just because the maintainer feels like being a dick.



  • @Helix said:

    I needed a windows FTP server

    Why did you not use the one bundled with windows?



  • Windows Server or Windows Desktop?

    I know there was an FTP daemon included with IIS under Win server. Didn't know there was an FTP server included in desktop/workstation windows.



  • @The_Assimilator said:

    Welcome to the world of open-source third party software, where a feature the creators of the software don't care about gets ignored, regardless of how useful or necessary said feature would be for users.

    FTFY



  •  @The_Assimilator said:

    @Helix said:

    I needed a windows FTP server just to quickly lash up a test - a quick google got me downloading Filezilla.

    I suppore the first suprise is that it didn't help me create a rule in windows firewall.  No problem, sorted that out.

    FTP Server is installed as a service, however every time i start the PC it asks to connect to the server admin console.  I had a quick look for an option to turn that off and I couldn't find one so a had a quick google to find the answer.  Only to find others asking the same question for the last 6 years and the following answer:

     

    Welcome to the world of open-source software, where a feature the creators of the software don't care about gets ignored, regardless of how useful or necessary said feature would be for users.

     

    i know this already,  i should have titled the thread as Open Source the existing name for crappy software

     



  • @Helix said:

    i know this already,  i should have titled the thread as Open Source the existing name for crappy software
     



  •  SAP is Open Source now?



  • Thanks for giving Blakey another fucking thread to rant on. I really thought this was his posting and was shocked to find otherwise.

    There are quality open source packages out there. Quality meaning that they're written reasonably well, will do the function intended, and will do it reliably and generally efficiently. We use Nagios for network monitoring. It may not fit your way of doing things, but for me it's light years above and beyond an application called Big Brother. (Yes, really -- put out by Quest Software.) We use an application called Cacti. It does what I need it to do and it does it well enough for me. Other packages that do the same functions (HP OpenView for link monitoring -- I don't know if it will also do SNMP data collection) I've heard from my colleagues are very difficult to configure correctly and cost thousands of dollars. I didn't have that problem here.

    You don't like what an open source developer is doing with the code? Fork it. Welcome to Icinga, which has had some features added to the original Nagios code. In my environment they'd be nice to have but not essential, so I've made the conscious choice to stay with Nagios.

    I would suspect there are lots of open source packages that were put out with a mentality of "I-had-a-need-wrote-this-quick-it-was-useful-maybe-it'll-solve-your-problem-too" and it just dies on the vine. There's really nothing wrong with that. You just need to know how to navigate the world of open-source software. Just like you need to know how to select an application in the world of commercial software.

    If you don't want to learn that, don't use open source software. Nothing says you have to.



  • @nonpartisan said:

    Thanks for giving Blakey another fucking thread to rant on starting a thread.

     

    edit: Really?  the strikethrough shows up in the "write a post" page, but not on the main page?  Screw you CS.

     



  • @nonpartisan said:

    You don't like what an open source developer is doing with the code? Fork it.
     

     Glaswegian accent needed


     



  • I agree it's an inconvenience but really, you get what you pay for. This is a very good product but it's free so you pretty much have to accept the limitations that the developer is not willing to address.

    In the case, it runs as a service, services are not meant to be turned off or on by switch preferences. it doesn't seem too unreasonable to just disable the service and re-enable it when needed...

    But this is just my opinion.



  • Who told you "open source" implies quality? You were misinformed.

    What it does imply is that you could take the source and fix the problems yourself. Legally.

    Try that with closed source software.



  •  Your main problem is that it doesn't have a shiny menu to switch whether it starts with Windows? Here's your shiny menu: services.msc

    If you don't know how to use windows services then you shouldn't have told it to install as a service



  • It is kind of unfortunate how hard it sometimes is to tell different kinds of open source apart.

    Generally speaking, one would hope thata team, even a team of unpaid volunteers, who are developing open source software to try to fill a need for a community in some respect...would do at least a little better job of documentation and responding to bugs and feature requests than one guy who wrote something for his own purposes, thought it was pretty useful, slapped an open source license on it and said "use at your own risk."

    Unfortunately, that's not always the case. Perhaps not even usually.



  • @Zolcos said:

     Your main problem is that it doesn't have a shiny menu to switch whether it starts with Windows? Here's your shiny menu: services.msc

    If you don't know how to use windows services then you shouldn't have told it to install as a service

    Way to completely miss the point.


  • @kilroo said:

    Generally speaking, one would hope thata team, even a team of unpaid volunteers, who are developing open source software to try to fill a need for a community in some respect...would do at least a little better job of documentation and responding to bugs and feature requests than one guy who wrote something for his own purposes, thought it was pretty useful, slapped an open source license on it and said "use at your own risk."

    The same thing could be said for many companies: you'd have thought a team of developers, all interviewed and paid to do a specific job, would produce something of quality and value that makes the company profitable... but many tales on this site shows that money and management don't necessarily equal value and quality.

    In terms of the open-source movement, they're a community like any other. They're subject to the same success and failures associated with good and bad management, they work within the boundaries imposed by policies and standards, if such exist - there are many free and open-source projects that put paid/closed-source stuff to shame, and vice versa. It boils down to the people driving and overseeing the entire process, be it open or closed source, driving effort into where it's needed the most.



  • @C-Octothorpe said:

    @Zolcos said:

     Your main problem is that it doesn't have a shiny menu to switch whether it starts with Windows? Here's your shiny menu: services.msc

    If you don't know how to use windows services then you shouldn't have told it to install as a service

    Way to completely miss the point.

    That neither the OP nor the person who answered that ticket understand windows services?



  • @Thuktun said:

    Who told you "open source" implies quality? You were misinformed.

    What it does imply is that you could take the source and fix the problems yourself. Legally.

    Try that with closed source software.

    That is exactly it. Open source software and closed source software are all just as bad quality. Open source softwares can potentially get fixed more quickly. Even open source softwares can be commercial softwares which can also result in good quality of supports and so on. (You can also find good quality softwares whether open or closed.)



  • You mean you got something for free and now you're complaining about it? I bet you download music and then forward a letter to the record producers telling them it sucked..

    With OSS, at least you have a chance to see why it is crappy, and you have a chance to fix it. (if you're competent)

    In fact that's one of the major promotions to open source software.

    Can you look at MS Word, and fix the bullshit that comes with that? GL, see you in court.

    Count your blessings, everyone's not the same, stfu.

    Three cheers for OSS.



  • @horroricane said:

    You mean you got something for free and now you're complaining about it? I bet you download music and then forward a letter to the record producers telling them it sucked..

    With OSS, at least you have a chance to see why it is crappy, and you have a chance to fix it. (if you're competent)

    In fact that's one of the major promotions to open source software.

    Can you look at MS Word, and fix the bullshit that comes with that? GL, see you in court.

    Count your blessings, everyone's not the same, stfu.

    Three cheers for OSS.

    Troll of the year award?



  • If that standard gains an award, I am disappoint. Not enough effort in the bait.

    Damn, I remember the days when trolls employed multiple complex techniques to really get under your skin; they were really quite brillant at it.

    Now it just seems to be a forgettable sneer in passing. Am I getting old?



  • @Sutherlands said:

    Troll of the year award?

    You kidding? That was pathetic. That wouldn't even pass muster at Slashdot.



  • @Sutherlands said:

    Troll of the year award?
     

    A post so useless, not even CS admits its existence:



  • @Sutherlands said:

    ]Troll of the year award?

    And why do you classify him as a troll?

    I just recently troubleshot a problem with an open source package (Cricket -- a predecessor to Cacti). I was able to find where the problem was and upgrade a library it needed. Had this been a commercial, closed-source product I'd never have been able to trace the flow and determine that a library dependency (in a package called RRDTool) had a problem with double-freeing memory with certain input values.

    Classify him a troll if you want, but he speaks the truth.



  • @nonpartisan said:

    @Sutherlands said:
    ]Troll of the year award?
    And why do you classify him as a troll?

    I just recently troubleshot a problem with an open source package (Cricket -- a predecessor to Cacti). I was able to find where the problem was and upgrade a library it needed. Had this been a commercial, closed-source product I'd never have been able to trace the flow and determine that a library dependency (in a package called RRDTool) had a problem with double-freeing memory with certain input values.

    Classify him a troll if you want, but he speaks the truth.

    Saying one true fact does not keep you from troll status.  This should be obvious to all but an idiot.


  • Jesus fuck, now I understand why blakey gets so uptight.

    All of you douchebags saying "if you have a problem with the source or maintainer, just fork it" are MISSING THE FUCKING POINT. I don't WANT to fork a shitty FOSS application, and be responsible in perpetuity for the resulting forked source, and have to periodically diff and merge in changes from the original project. I just want my small change that makes the application better, to be merged into the main trunk so that I can download a binary from the project's official site and know it has that change in it.

    The problem with FOSS is not the concept or the code, it's the people who run the projects and the culture of "I made this app so I am God of it". Bull fucking shit. If you put an app on the Internet, it's because you expect people to use it. If you expect people to use it, you'd better fucking well be prepared to accept criticism and suggestions from those people. And if you want them to continue using your app, you'd better fucking well DO SOMETHING about the criticism and suggestions you get.



  • @The_Assimilator said:

    The problem with FOSS is not the concept or the code, it's the people who run the projects and the culture of "I made this app so I am God of it".
     

    That. And even then, the same issues affect non-FOSS stuff, too.

    See "Spectate Swamp Desktop Spagetti Golf Random Noodle" for further details.



  • You want the point? Okay, here's the point.


    The point is that open source software gives you options. It gives you the possibility of getting something up and running quickly and cheaply. Outside of the work environment, it gives you the ability to get a service online that perhaps you wouldn't do otherwise due to cost. If I'm going to host my own Web server at home, I sure as fuck am not purchasing IIS for what amounts to a hobbyist setup. What'll I put up? Apache. It's well known, it's supported, and the only cost is whatever it takes to get it installed and configured.


    If you're in a crunch and you need a quick FTP server, download an FTP server. Install it. Get it up and running. Don't purchase additional licenses or what-have-you for other products.


    Are there good open source FTP servers? Absolutely. Are there shitty ones that aren't worth the physical space being taken up on the drive? Damn straight.


    I read about people railing against Lotus Notes, saying it's one shitty product. I've never used. I have in the past used GroupWise. In some ways, GroupWise was much better than Outlook. In other ways, not so much. But you can get a shitty commercial product or you can get a superior commercial product. Gee, that sounds like open source too.


    Open source doesn't automatically mean quality. Neither does a commercial product. Commercial products are generally better supported because they're bringing in income and so they must support it. Open source products, not necessarily the same. In other words, you need to do research in the open source area just as much as you need to do research in the commercial area. But there are several well-known products in the open-source arena that I would jump to as a starting point if I needed something:

    • Web server: Apache
    • FTP server: ProFTPd
    • SSH server: OpenSSH
    • DNS: BIND
    • Mail server: sendmail or qmail
    • DHCP server: ISC DHCP

    Open source isn't rocket science. And if you just don't fucking like open source, then just don't fucking use it. Open source has bugs. No one's saying it doesn't. You need to know and understand the benefits and limitations of open source just like you need to know the benefits and limitations of the commercial products you're trying to get into.

    Open source does have the benefit of being open. I have seen that myself where I've been able to fix bugs on my own because the source was available. Do I do it all the time? No. Is it the main reason I use open source? Hell no. I use open source because it's convenient and I'm comfortable working in the open source paradigm. If you don't like it, don't fucking use it. Period.

    THAT, my friends, is THE FUCKING POINT.



  • C-



  • @The_Assimilator said:

    Jesus fuck, now I understand why blakey gets so uptight.

    All of you douchebags saying "if you have a problem with the source or maintainer, just fork it" are MISSING THE FUCKING POINT. I don't WANT to fork a shitty FOSS application, and be responsible in perpetuity for the resulting forked source, and have to periodically diff and merge in changes from the original project. I just want my small change that makes the application better, to be merged into the main trunk so that I can download a binary from the project's official site and know it has that change in it.

    The problem with FOSS is not the concept or the code, it's the people who run the projects and the culture of "I made this app so I am God of it". Bull fucking shit. If you put an app on the Internet, it's because you expect people to use it. If you expect people to use it, you'd better fucking well be prepared to accept criticism and suggestions from those people. And if you want them to continue using your app, you'd better fucking well DO SOMETHING about the criticism and suggestions you get.

     

    Now you're succumbing to the same problem.  The person who wrote the app is the god of it.  Period.  If they decide to update it, they're nice.  If you've purchased the app, and they don't update it, then the app was probably not worth the money.  If the app was free, than you literally have no grounds to complain.  You did not pay for it.  And I can tell you from experience, a lot of OSS developers put their app on the web because someone else might find it useful, but they've moved on, and don't care about it anymore.

     



  • @GFreeman said:

    I agree it's an inconvenience but really, you get what you pay for. This is a very good product but it's free so you pretty much have to accept the limitations that the developer is not willing to address.

    In the case, it runs as a service, services are not meant to be turned off or on by switch preferences. it doesn't seem too unreasonable to just disable the service and re-enable it when needed...

    But this is just my opinion.

     

     

    No, it's not about turning the service on and off - it's about being about to simply enable/disable the admin console on startup without having to edit the registry.



  • @Helix said:

     

    i know this already,  i should have titled the thread as Open Source the existing name for crappy software

     

     

    I think i might re-trim the title to "Open Source the existing name for dickweed maintainers who prefer to argue then allow an app to mature".

    I am suprised no one has mentioned how open source is like a car that you change the engine... 

     



  • @Helix said:

    I think i might re-trim the title to "Open Source the existing name for dickweed maintainers who prefer to argue then allow an app to mature".

    You still don't get the points people are making in this thread.



  •  So someone creates some software and published it, for free. Then you complain it doesn't have certain features you want it to have. It's open source, you could adjust it if you liked. Or you could pay money for a commercial product which has features you want and support. I see no WTF here.



  • One of the issues with Open Source Software is that the "Open Source" part is used both as a hook and as an excuse. "You can't fix problems in commercial software" well, no. But when you tell the developers about it, they'll generally fix it, rather then tell you to fix it yourself, or bitch about how the people are complaining they got something for free. It's still a product, and any sane person is going to move on to another product if it doesn't meet their needs. I don't care if they get paid for what they do. I don't care if they think they are fulfilling some noble cause by providing their often-broken shit on the internet for free. I care whether I get something that works. if it doesn't work, I find something else. I don't dink around in other peoples code to fix their stupid fucking mistakes because they are too fucking important to do it themselves. useful features? Those I can understand being added by others. But having to fix shit that is already broken? That's ridiculous.



  • @Cassidy said:

    @Helix said:

    I think i might re-trim the title to "Open Source the existing name for dickweed maintainers who prefer to argue then allow an app to mature".

    You still don't get the points people are making in this thread.

     

     

    Yes I do.  Which ones do you think i do not get?

     



  • @Shishire said:

    If the app was free, than you literally have no grounds to complain.  You did not pay for it.
     

    It's high time this sentiment was killed. It is the dishonorable low road of software developmentand and lowers standards across the board.

    Whatsoever man builds, that he shall also support.




  • @Helix said:

    Yes I do.  Which ones do you think i do not get?

    The fact that software can be categorised in many ways:

    • good v bad
    • open-source v closed-source
    • free v paid-for 

    Hence any package can fit a characteristic from all three list items there, for example:

    • it could be closed, paid-for but shit
    • it could be free, work well - but still be closed-source (author wants to retain IP for their code)
    • it could be open-source, good and free
    • it could be open-source, free but still shit
    • etc...

    Reading your posts, your responses seem to be fixated upon the concept that "open-source" automatically implies "must be shit" or "developed by incompetants" and dismiss all the information that has gone on before, as though you have fingers firmly stuck in your ears and refuse to listen to attempts to broaden your viewpoint.

    I don't know if this stems from the fact that you feel you've been hard-done by a single open-source application and have decided to judge them all in that same way, or if you've genuinely got extensive experience of open-source utilities and various O/S communities and managed not only locate just the utilities that suck badly, but also choosen development groups famous for their infighting, ineptness and lack of vision - but I honestly feel you're not only being judgemental based upon your experiences you've posted here but also stubbornly ignoring posts from other users trying to explain concepts behing open source, and their related success stories.

    Or have I just been trolled? I'm probably getting too old to spot it, you know...



  • @dhromed said:

    @Shishire said:

    If the app was free, than you literally have no grounds to complain.  You did not pay for it.
     

    It's high time this sentiment was killed. It is the dishonorable low road of software development and and lowers standards across the board.

    I also disagree with that statement, simply because it implies an author of free software could dismiss all complaints and live in their deluded bubble that their application was perfect and never needs tweaks nor improvements, leading to the bigger picture that the developer always produces high-quality code - there is no motivation or desire to improve or learn alternative ways of doing things. I mean, just look at Spec Swamp.

    I also feel that I do not need to pay for an app in order to complain about it - perhaps one of my complaints about its functionality during a demonstration and the way that complaint is addressed is the reason why I refuse to purchase that app in the first place.

    I will concur that - being as I paid nothing and the author received nothing - that the author is under no obligation to listen to my complaint, nor act upon it. However, most good developers view complaints and criticism as improvement opportunities and actively invite feedback about issues that need addressing.

    (How those reports are formulated and collated is another discussion - but I'll have to say that Blakey's bug report was an excellent example of providing valuable information.)

    @dhromed said:

    Whatsoever man builds, that he shall also support.

    Well, that he is accountable for supporting it - he may have moved on and performed a formal handover to a community that have now taken ownership of that product. Partly the success behind Linux, cacti, nagios, postfix, apache, wordpress etc...  Either way, if he started it, he should "finish it" in some way.



  • @Cassidy said:

    Hence any package can fit a characteristic from all three list items there, for example:

    • it could be closed, paid-for but shit
    • it could be free, work well - but still be closed-source (author wants to retain IP for their code)
    • it could be open-source, good and free
    • it could be open-source, free but still shit
    • etc...
    Could be != is. In reality, OSS is unimaginably shitty software in almost all cases. Let's not beat around the bush here: it's simply because most developers who are 'comfortable with the OS paradigm' are really fucking stupid, so their code and design practices are bloody awful. We all know it's true. Apart from a small percentage of very bright hippies/neo-communists, the only people who write OSS are those who can't get a proper job because they can't code Towers of Hanoi.



  • There are a few good open source products, but they invariably have corporate governance and work under corporate management structures. None of the ad-hoc open source products on SourceForge are worth shit, and frankly even the governed products (like, from the Apache organization) are mostly crap.

    Mozilla is an interesting example because, while it fits the above example, the corporation was actually created purely to write open source software. In all other cases, the corporation already has a track record of writing closed source software, and moves into open source via either purchasing it (Oracle), or deciding they don't want to rely on other people to advance their goals (Google's Chrome).

    And note that while Chrome is one of the few good examples of open source software, it still has a lot of usability faults (especially in the dev tools), the developers still don't read their own fucking bug tracker, and it's still hated by most of the open source community because they're paranoid Google is spying on them.

    Let's talk about "being free" for a moment-- earlier in the thread someone made the claim that, since Windows/IIS costs money they're better off using Apache to get the job done. Well, two main points here:
    1) You lose a lot of opportunity-cost getting Apache installed and configured correctly for any non-trivial site. Just installing a SSL cert is a nightmare of painful UI that, if you weren't already an Apache expert, you'd have to spend days working through. (In IIS? It's a 4-step wizard.)
    2) We all live in 2012. You can read a Windows cloud server for about $250 a month. (Actually, Amazon even has a free version which has a wimpy CPU, but is fine for developing your initial product, and it's free. Free free, not freedom free.) At $250/month, the cost of spending 3 days learning how to install a SSL cert into Apache when you could have instead used that time developing your product, it suddenly becomes a waste of money.

    This is what Microsoft means when they talk about TCO, "Total Cost of Ownership", which isn't simply the licensing cost, but also considers the payroll costs of the product.

    Open source products beat Microsoft products in terms of TCO if, and only if, your staff is already trained in using and maintaining them-- meaning you offloaded that cost to some other company and just made a savvy hiring decision, or you hired some complete mega-nerd to thinks learning how to install SSL certs into Apache is "fun". And guess what? Even then the TCO argument goes away when your knowledgeable employee leaves for greener pastures, and you realize you have to pay $30k/year more to hire another one (over hiring someone who knows only the Microsoft version.)

    Anyway, yes, there is a lot of closed source software that sucks ass. One of those companies writing suck-ass software now owns two high-profile open source products (Oracle, OpenOffice and MySQL respectively), so suck it down! Oh, and BTW: it's hilarious to me when open source proponents praise IBM for supporting and using open source products. Have you ever SEEN any IBM software? Christ, it's so buggy, bloated and convoluted it makes Oracle look good by comparison. (Lotus Notes is the poster-child for this, of course, but it's far from the only offender.) Getting endorsements from IBM and Oracle? That is not a boon.



  • @MascarponeRun said:

    In reality, OSS is unimaginably shitty software in almost all cases. Let's not beat around the bush here: it's simply because most developers who are 'comfortable with the OS paradigm' are really fucking stupid, so their code and design practices are bloody awful. We all know it's true.

    Again, I don't know why people are fixated on this idea. I use a lot of OSS (apache, cacti, mysql, php, postfix, fail2ban, awstats, iptables, xbmc, wordpress etc) and although I'm not tub-thumping about how utterly brilliant, idiot-proof and noob-tolerant it all is, I find it no better or worse than many of the closed-source products. Perhaps I've just managed to luck on a subset of OSS that works fine and only been exposed to low-quality proprietry stuff that the standards are close, but it's highly unlikely.

    @MascarponeRun said:

    Apart from a small percentage of very bright hippies/neo-communists, the only people who write OSS are those who can't get a proper job because they can't code Towers of Hanoi.

    And yet I have read interviews with OSS authors that are developers/sysadmins/network engineers/project managers by day and volunteer their skillsets to assist in OSS development in their spare time, many of which are driven by a frustration of not being able to improve an existing closed-source tool so have taken it upon themselves to create their own. Granted, they're starting from existing use cases and someone else's concepts (thief!), but this seems bourne out of a rejection that things cannot change and people deserve better.

    Look, I accept that not all OSS is a shining beacon of efficiency and brilliance. But in the last 15 or so years I've run my own webservers, mail servers, FTP servers, IRC networks, media centres; I've automated a lot of routine tasks and been pretty productive with OSS products. I often get people telling me I can do the same under Windows, or that there's some other applications that can do it for me - but the point is that I can't do it under Windows (simply because I don't know how to) and advocates of those applications can't sell me on the benefits of using their suggested application over the OSS product I've already implemented. Perhaps I'm blinkered because I've spent so long in that environment, but one of the reasons OSS isn't adopted for more widespread use is this outdated notion that it's automatically crap.

     



  • @dhromed said:

    It's high time this sentiment was killed. It is the dishonorable low road of software developmentand and lowers standards across the board.

    Whatsoever man builds, that he shall also support.

     

    It really depends on the individual's motivations.

    Back in the BBS/Fidonet days, I was very interested in electronic communication.  I used terminal communications software constantly.  I was going through an assembly language bent at the time and decided I wanted to write a simple terminal communications program myself.  I never found any good explanation on what the general flow should be, so I wrote something, tweaked it, and it worked okay.  I had a pride for this simple app, founded or un as it may have been, and I wanted to share it.  I was a sysop and I had some friends who were sysops, so we all posted it.

    I don't recall if anyone ever downloaded it.  I left my e-mail address and Fidonet node number in the docs.  I never heard anything.

    Was it a piece of crap?  Yeah, probably.  Did it serve a purpose?  In my mind, it did.  It was the first terminal communications program I'd written.  I thought it may provide someone else a basis for writing their own or getting an idea of how such software worked.  Did I ever do anything with it later to support it?  Meh, not really.  I moved on to other interests (like writing my own Fidonet mailer in assembly -- I may even still have the source for that).

    People write and share their software for different reasons.  I suspect, but have no proof, that a lot of people who have software published on SourceForge write their software to be "good enough" for their purposes, want to share it, but then move on.  Maybe they used SourceForge because they had a collaboration going on between a couple of other developers with a common interest but they've finished what they were doing.  So if it suits your need, great.  If it only works on Windows, that was their need at the time.  If you need something similar under Linux or OS X or AIX or Apple DOS or whathaveyou, it's not going to be the piece of software you need.  So you need to find an alternative.

    You may consider it dishonorable.  On some level it may be.  And I've seen responses that vehemently object to the "fork it" mantra.  But if an individual finds that someone else's work can be the basis of their own project, even if that first project isn't maintained, well . . . is that really bad?  Especially if the result is a project that ends up being well-maintained and well-supported?  (I don't know much about NetSaint, but I know that was the basis for Nagios, which is now a well-supported monitoring system.  And I'm grateful for the beginnings that got it where it is today.)

     



  • @blakeyrat said:

    Let's talk about "being free" for a moment-- earlier in the thread someone made the claim that, since Windows/IIS costs money they're better off using Apache to get the job done. Well, two main points here:

    1) You lose a lot of opportunity-cost getting Apache installed and configured correctly for any non-trivial site. Just installing a SSL cert is a nightmare of painful UI that, if you weren't already an Apache expert, you'd have to spend days working through. (In IIS? It's a 4-step wizard.)

    2) We all live in 2012. You can read a Windows cloud server for about $250 a month. (Actually, Amazon even has a free version which has a wimpy CPU, but is fine for developing your initial product, and it's free. Free free, not freedom free.) At $250/month, the cost of spending 3 days learning how to install a SSL cert into Apache when you could have instead used that time developing your product, it suddenly becomes a waste of money.
     

    I have a theory, completely untested, that for all of the complaining people do about the learning curve of non-Microsoft vs Microsoft solutions, that in the end the learning curve ultimately evens out.  Microsoft products give you the ability to create something basic, something that looks reasonable, and do it quickly.  But when it comes to more advanced features, you end up spending the same amount of time as any non-Microsoft product.

    If I want to write a document, I can fire up Microsoft Word and I can get the basic document entered quickly with little or no training.  If I want to write the same document in TeX or LaTeX, not so much if I'm a newbie.  There's a steeper learning curve for using LaTeX.  But if I want to use more advanced features of Word (tables of contents, master documents, etc.), then I'm going to spend a significant amount of time reading help files, purchasing self-help books, taking classes, etc.  That's the hidden cost I believe most people don't think about.  And if I'm having to read help files, self-study guides, etc., what's the difference in the time spent between the two ways of doing it?  I've had Microsoft Word fuck up my documents so badly that I had to copy and paste the original, in sections, into a brand new document.  (Full disclosure:  admittedly it has been a long time since this has happened -- Word 97 maybe.  But as a result I don't have it automatically generate tables of contents, or use master documents, or any of those features for fear that it is going to get all screwed up like it did.  But how about the time wasted there?)  LaTeX isn't for everyone.  It's a markup language.  But when something doesn't look right, having learned the LaTeX language structure, I can use that to figure out why my document doesn't look right.  That was one of the huge advantages of WordPerfect -- Reveal Codes.  You could see what was interfering, delete a "Bold" opcode (or whatever) where it didn't apply, and your document returned to looking the way it should.  It's not the same under Word.  But when it comes time for using the more advanced capabilities, I suspect the total time spent in trying to learn either method would be a wash.  (That's why I thought it was so damn hypocritical of Microsoft to say that there would be such a steep learning curve if people switched from Office 2003 over to OpenOffice -- which had pretty minor differences in UI structure -- then lo and behold they completely and totally change the interface for Office 2007 and say users would pick up on it right away and there would be minimal cost in the change.)

    Getting back to Apache, you know that installing a certificate is a four-step wizard.  Did you automatically know that?  How did you find that out?  How many times did you try it and find that it didn't work exactly the way you wanted?  Maybe it didn't associate the certificate with the right site.  I'm just guessing, as I have never used IIS.  Did you have to look in a help file?  Did someone who has experience with IIS tell you how to do it?  These are all factors that would affect hidden costs of "it's a four-step wizard."  Perhaps still not as long as it takes to learn how to do it in Apache, but likely longer than the statement implies.  FWIW, I'm searching for '"install a server certificate" iis 7.0' and not finding it to be that simple.  I'm reading that I need to [url="http://technet.microsoft.com/en-us/library/cc732785(WS.10).aspx"]import the certificate before I can assign it in IIS.[/url]  For someone who has never used IIS for a secure site before, this doesn't sound like a four-click wizard.

    Regardless, in both IIS and Apache, once you've performed the function multiple times it gets simpler each time to the point where the extra time taken the first time is negligible.  (Think of a function evening out as it goes to infinity -- the initial values are eclipsed by what happens in the distance.)  But Helix originally spoke of needing an FTP server quickly.  Basic FTP is not a secure connection.  He needed something quick.  So getting back to Apache, if you wanted to install Apache quickly, without any bells and whistles, you could do it without having to install a certificate.  I could throw up an open source OS (FreeBSD, Linux) and tell it to install the Apache package and the basic installation is done.  But even under Windows, I just looked on the Apache Web site and the mirrors even have a binary Windows .MSI for version 2.0.64 that you could install.

    All of this said . . . I did a search for "how to install apache cert" and came up with:  [url="http://www.digicert.com/ssl-certificate-installation-apache.htm"]this[/url], [url="http://www.geocerts.com/install/apache_2"]this[/url], and even [url="http://slacksite.com/apache/certificate.php"]this one that talks about creating a self-signed certificate[/url].  I don't believe it would take three days to get a certificate installed under Apache.  And if you're Googling for "why didn't this work under IIS" if you run into a problem, you'll probably find as much help Googling for "why didn't this work under Apache" too.



  • @nonpartisan said:

    Microsoft products give you the ability to create something basic, something that looks reasonable, and do it quickly.  But when it comes to more advanced features, you end up spending the same amount of time as any non-Microsoft product.

    99.999% of all products never need "advanced features." So while your theory may be correct (and I doubt it is), it's still not an argument for using (say) Apache over IIS in the general case.

    @nonpartisan said:

    (Full disclosure: admittedly it has been a long time since this has happened -- Word 97 maybe. But as a result I don't have it automatically generate tables of contents, or use master documents, or any of those features for fear that it is going to get all screwed up like it did. But how about the time wasted there?)

    So:

    a) You using the annoying-as-fuck open source tactic of comparing the current version of an open source tool to a decades-old version of the Microsoft tool

    b) You are an idiot who doesn't use backups (nobody's going to say that Word 97 didn't corrupt documents occasionally, although my my experience 100% of the time that happened the document was sitting on a bad HD block, but you didn't keep backups? In 1997 when all software and hardware was 50 times less reliable than today? On a Fat32-formatted disk? Idiot.)

    c) You're blaming us for the time wasted by your mental illness that causes you to refuse to use Word's time-saving features?

    For just once, can I debate with someone who thinks rationally? Just once I want to have a debate like this without the opponent being completely delusional!

    @nonpartisan said:

    That was one of the huge advantages of WordPerfect -- Reveal Codes.

    Word is style-based. It also lets you select-by-style. Those features let you fix a document a thousand times faster than using reveal codes and scanning through the entire thing to find rogue "Bold" codes. Of course you don't know anything about Word, obviously, but I thought I'd point that out.

    @nonpartisan said:

    That's why I thought it was so damn hypocritical of Microsoft to say that there would be such a steep learning curve if people switched from Office 2003 over to OpenOffice

    Cite? When did Microsoft say that?

    @nonpartisan said:

    then lo and behold they completely and totally change the interface for Office 2007 and say users would pick up on it right away and there would be minimal cost in the change.

    Except users (at least those ones who actually tried it instead of bitching and moaning about it) did pick it up right away. New users picked it up every quicker. Both classes of users are much more efficient using the Office 2007/2010 interface over the Office 2003 interface. How do I know? Because Microsoft actually does usability studies, and changes their product based on the results. They're also courageous enough to say, "fuck this" to the "we've always done it this way" attitude in their most important product which, even if you hate the Ribbon interface, you have to admit is pretty admirable and gutsy.

    @nonpartisan said:

    Did you automatically know that? How did you find that out?

    The company that sold me the SSL cert sent detailed instructions on how to install them in IIS. They didn't even bother to attempt to write up instructions for Apache.

    @nonpartisan said:

    How many times did you try it and find that it didn't work exactly the way you wanted?

    Zero.

    @nonpartisan said:

    I'm just guessing, as I have never used IIS.

    Obviously.

    @nonpartisan said:

    I'm reading that I need to import the certificate before I can assign it in IIS.

    You can actually do that by double-clicking it.

    @nonpartisan said:

    Regardless, in both IIS and Apache, once you've performed the function multiple times it gets simpler each time to the point where the extra time taken the first time is negligible.

    The vast, vast majority of people install a SSL cert once. Ever. Well, ok, once per product. Still, it's not something that you do five times a week, so the extra time taken to learn it is certainly not negligible.

    Your rambling tirade of ignorance there just isn't doing it for me. Maybe try actually using Office 2007, or IIS before talking about them would help? Maybe actually thinking about what you're saying ("herp derp people install SSL certs 10 times a week derp derp") before saying it would help? Maybe... less pointless filter? Shorter, more readable, paragraphs?


Log in to reply
 

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.