Feedback, In Software as Well as In Life, Gets Results

A video documenting a simple problem has gone viral on Youtube. One step from a subway station is a tiny bit higher than others and people following the rhythm of the stairway often trip on it. Bad things Happen.
Good things happen, too, when ordinary folks, users, tell service providers what they think of the service. Sometimes one picture is worth a thousand words. Other times thousands of responses get the attention of people in a position to make the world a better place.

In software, the same principle applies. The people who make software are not perfect. They make mistakes. They misunderstand things. They get priorities wrong sometimes. They don’t know every situation in which software will be used.

In FLOSS there are often well-defined means of giving developers feedback. Sometimes e-mail works. Other times a bug-tracking system works. FLOSS also has a backup system for when feedback fails to get a better result, the fork. Since the code is open for examination and modification, it is feasible for a user or group of users to fix the problem or to improve the software. It is even possible to make informed decisions about starting from scratch to replace software. Choice is good.

FLOSS beats closed source software in many ways thanks to the openness of the source-code. Often there is no way to ensure feedback gets to the people in charge with non-free software. There are just too many layers involved. I remember the time I was working in a school and several times reported on paper a problem with the configuration for printing. My students, using GNU/Linux could print but no one else in the building who was a student could print. At first, I was told I used the wrong form. On the right form, nothing happened. Finally, I physically assaulted an IT person and demanded to know why the problem had not been fixed. “What problem?” was the reply. It turned out that my boss had rewritten my report. The secretary in the IT department also rewrote it. By the time the IT guys got it, the report was watered down to “Hello, World!” or something so generic no action need be taken. It took the IT guy seconds to allow students to print…

So, if you want your feedback to count, if you want to use software that increasingly works for you and not someone else, use FLOSS, the right way to do IT. I recommend Debian GNU/Linux because it works for you. Debian has a very effective bug-tracking system. Of course, there is a package for that and you can add it to any computer running Debian GNU/Linux. How’s that for openness? You are not restricted to using that package for reporting software bug either. You could use it to report on stairs, building maintenance, gardening chores, “Honey, Do” lists and so on.

About Robert Pogson

I am a retired teacher in Canada. I taught in the subject areas where I have worked for almost forty years: maths, physics, chemistry and computers. I love hunting, fishing, picking berries and mushrooms, too.
This entry was posted in technology. Bookmark the permalink.

18 Responses to Feedback, In Software as Well as In Life, Gets Results

  1. oiaohm says:

    Chris Weig XP added more issues than 2000. Secuirty wise 2000 should be put out of its misery. Problem is XP, Vista and 7 really don’t match the stability of 2000.

    This is why I talk about MS going backwards in security and stability. 2000 does not have a highly complex interface running. XP, Vista and 7 are running a more complex interface than 2000. Windows 8 if enough new apps appeared would see stability improvement.

    Chris Weig
    “The beast was running Linux, and restarting the machine took an excruciating 3 minutes and 11 seconds. (I’m not making this up, it happened.)”
    Embedded and I can tell you why. Less than 1 second start is possible but when it comes to reverse vending machine error triggers a full memory check and storage check before returning to operation.

    If bad ram was found in Linux any failed section of ram would have been zoned out.

    No one worked out a magically fast way to check harddrives or ram for defects in software.

    Lot of Linux cash vending machines have the same policy fail check ram so system cannot fail again due to ram fault. Yes due to the rarity of random crashes in Linux the system goes looking for broken hardware.

    There is a fast way to check ram for defects but it requires spending more on the controller board. Lets just say those reverse vending machines are built on the cheap. The fast way is used with Linux in aircraft cockpits. 1 second is possible if the right hardware is used. Some coke a cola drink machines running Linux have aircraft grade in them. So you would not even see a crash because the system is back online in 1 second. You see a blink that is it.

    Chris Weig so yes that 3 mins and 11 seconds is more annoying for people like me who know that only happens because the controller board is cheep. The price difference between a fast start controller board for Linux and snail from hell board is 20 USD. That 20 USD adds hardware checking of ram and hardware protection for data storage and a directly connected flash in memory space so Linux can fire without having to load anything into ram.

    Mind you 3 min and 11 seconds is no the worst I have seen. Windows based cash machine made by diabolic same ones who made voting machine that could not add up. Power outage boots up fine now this is the big but the card reader has not fired up. You insert you card it keeps it and asked you to type in the number off the face of card because the reader is broken. Why can you not its still in the machine.

    Get worst after so long not entering number takes card places in internal box goes back to looking like its operational again waiting for next victim.

    Mind you I would not say this is a Windows or a Linux only problem. Miss hardware check bad things happen to end users.

    Yes booting up fast not doing a hardware check creates some serous problems far worse than having people wait. Don’t worry about having to wait 3 min and 11 seconds it did not nick your keycard or not pay you. Yes it worse if these machines come on-line and not be functional can be way worse particularly when they are like some of the diabolic brand devices. Yes diabolic is a real brand of cash machines. You would think the company name is a hint.

    The Linux machine could have gone 3 mins and a 11 seconds then displayed out of service if it found something serous-ally wrong that Linux cannot patch over.

  2. Chris Weig says:

    I don’t believe that is true unless you haven’t been using that other OS.

    Your choice. I’ve got no reason to make this up.

    A relative of mine is running a transport company, and their office computers (about 8 years old) are running with Windows 2000. Not the wisest of choices security-wise, but the things run and run and run without ever crashing. Yes, with a Microsoft OS.

  3. Chris Weig wrote, “I haven’t experienced a single Windows crash since Windows 2000.”

    I don’t believe that is true unless you haven’t been using that other OS. I was at a school in 2004 where I could crash that other OS in minutes by opening more than a few windows in the browser and firing up the word-processor. That was Lose ’98. I have seen XP crash just idling after a few days. That was early on. M$ has had the opportunity to fix a lot of bugs.

    “7” was crashed with a simple HTML tag…

    A recent presentation of “Surface” also suffered a freeze.

  4. Chris Weig says:

    Come on, Mr. Pogson. These clips of Windows 98 crashing are getting very, very old. I haven’t experienced a single Windows crash since Windows 2000. With other people’s computers, crashes happened in almost all cases because of faulty 3rd party drivers.

    Funny side note: yesterday I experienced a crash of my supermarket’s reverse vending machine. The beast was running Linux, and restarting the machine took an excruciating 3 minutes and 11 seconds. (I’m not making this up, it happened.)

  5. Clarence Moon wrote, “The facts of the matter are that Windows does not fall on its face with any such regularity, particularly if it is under the care of some competent admin, which you claim to be.”

    I suppose this was just comic relief?
    http://www.youtube.com/watch?v=P38cLZJ0w2s

    75% or so of PCs running that other OS have only a consumer for system administrator. You cannot blame them for out-of-the-box performance.

  6. Clarence Moon says:

    Many people find it simpler and cheaper to buy a new PC than to fix the old one.

    Hand-waving at best, Mr. Pogson. Hundreds of millions of copies of Windows are sold annually and way more than a billion are currently in service. You sneer at that and claim some kind of superiority for Linux, but not one in a hundred buy a computer with Linux. Hobbyists may raid the junk boxes and trashcans of the world and install Linux and proclaim themselves to be genius material, but they are only fooling themselves.

    And you change the topic from Microsoft abandoning their responsibilities to maintain their products to some vague, unsubstantiated claim about something that you were administrating falling on its face with some regularity. The facts of the matter are that Windows does not fall on its face with any such regularity, particularly if it is under the care of some competent admin, which you claim to be. I think you are grossly exaggerating your claims and failing to see the sort of light you are painting yourself with.

  7. oldman says:

    “One cannot expect a developer to maintain his previous work indefinitely any more than Ford maintains Model Ts.”

    And yet this is precisely what you rake microsoft over the coals for. Of course we now know the real problem IMHO you have with Microsoft and their ISV’s is that they wont work for free.

    “The only reason that I switched to GNU/Linux in 2000 was that other OS going “Poof!” in every class, all day long.”

    I’ve always wondered Pog…

    What other OS Pog? windows 9x? running on what? How well was it maintained?

  8. Clarence Moon wrote, of that other OS going “Poof!”, “that has never happened”.

    Millions of copies of that other OS have been sold to replace ones that did drown in malware or refuse to boot. Many people find it simpler and cheaper to buy a new PC than to fix the old one. In fact, many know that that other OS is defective by design and cannot be fixed yet they know a new PC is likely to run for a time.

    The only reason that I switched to GNU/Linux in 2000 was that other OS going “Poof!” in every class, all day long.

    I was in a place in 2010 where “fixing a PC” meant putting the box on a plane, hiring a taxi in Winnipeg to take it to a fix-it shop, and reversing the process to get it back, probably $300 for a round-trip. New PCs can be bought for $300 and are likely to be faster than the year-old machine that may or may not be fixed. After I left, everyone in that community knew that a PC with that other OS can usually be fixed by having one of my students install Debian GNU/Linux for the cost of a can of soda pop. The only wrinkle we found was getting a LoseModem to work. We found a $20 USB modem worked out of the box so there was a brisk business importing those. $300 worth of expense replaced by $30 worth of expense for GNU/Linux was clearly a better solution. Faster, too, if the high-speed Internet connection was used instead of dial-up.

  9. ch wrote, “Note the “honeypot” part?”

    That’s his opinion. They still have a bug-tracking system apart from “Brainstorm”, so, what’s your point? Users still provide feedback. Not every developer is involved in quality control. Those that are will pay attention to what users are saying. It’s a matter of scale. Not every developer will read what every user writes about every package. There needs to be filtering and that’s what the bug-tracking systems do. In Debian’s, for instance, reports get categorized every which way and the right people get to say or do what they need to improve the software. One of their options is to report bugs “upstream” to the folks who write the code. It works. Stuff that’s a matter of documentation or configuration is often fixed by Debian package managers while actual code may be fixed by the original authours. This is a natural way to do things as there may be a million users of one piece of code and developers have not the time, energy or resources to chat with them all. M$ does not seem to even try to connect users with developers in any way. It’s always the next release kind of loose handshake, if anything. In all my years of using that other OS, I never met a user who complained to M$ about the software because it was pointless to do so. If M$ shipped software known to crash and burn and suck up malware, what chance did a normal bug have of getting their attention?

  10. Clarence Moon says:

    after a year it can go POOF! after a year it can go POOF!

    In a legal sense, perhaps, but that has never happened and that FACT is what people rely on regarding fixes to Windows OS problems. No such confidence exists for FLOSS projects that can blow away with the next summer breeze. No one has any skin in the game and that makes it easy for them to simply quit when the going gets tough. You cited one such case here within the past week.

  11. oiaohm says:

    ch yes I work around the wine project. I am not a developer. I am in the team of support people. Some do bug cleaning yet are not developers others handle the forums working out what posts need to be sent to developers.

    So yes in wine developers do answer end user request directly when the fault is proven even in the wine forums or mailing list.

    Ubuntu theory about putting barriers between end users and developers is a foolish one. Yes wine we have a filtering process. But if you bug passes the developer talks in email irc …. to the user suffering the problem so the problem is exactly understood so it is fixed.

    Usability is another matter completely.

    http://www.openusability.org/ exists to provide audit teams to turn all those opinions into to mock ups and see if the idea has legs.

    https://launchpad.net/ayatana from ubuntu is a classic example of not being willing to work upstream. So creating a stack of hacks that cause applications to malfunction on ubuntu where they don’t malfunction on any other distribution. Really ayatana is not how to fix usability its how to break it.

    Wine did not require to set up a honey pot to draw noise away from bug tracker. In fact ubuntu duplicate bug tracker and shipping patched versions of wine is annoying.

    Noise in the bug tracker is important to know how many people are hitting particular bugs to give bugs order or critical to fix.

    ch ubuntu is gulity of doing lots of things wrong. Problem is people blame all of Linux for faults Ubuntu has not seeing it is items like ayatana and other cases of not working with upstream causing them.

  12. ch says:

    Mr Pogson,

    two years ago the issue of “feedback for FLOSS” was discussed here:
    http://piestar.net/2010/06/22/ayatana-missing-the-point/

    One of the participants was MPT (Matthew P. Thomas), who (at least at the time) worked with Canonical. Here’s an excerpt of what he had to say:
    (start first quote, 2010.06.26 02:37)
    I work for Canonical, but I am speaking strictly for myself here. Formerly I was in the Launchpad team, working on pages like that login page. Now I’m in the Design team, working on things like Ayatana.
    (end first quote)

    (start second quote, 2010.06.27 01:15)
    As you say, “There is plenty of user feedback coming at FOSS products” — but most of it is useless, because opinions are not data.

    Imagine you are an engineer, or even a designer, with eight hours to devote to Ubuntu today. What is the best way of spending that time? Will reading e-mail notifications of new bug reports help you use the rest of your time better enough to make it worth it? Maybe or maybe not, depending on the package. Will commenting on posts on Brainstorm, or Ubuntu Reddit, or OMG Ubuntu, or Piestar help you use the rest of your time better enough to make it worth it? Probably not (though it varies by post, hence me here). Will reading the Ubuntu Forums help you use the rest of your time better enough to make it worth it? Almost certainly not. Forum contributors, if they see a response from a developer, may be reassured or inspired to contribute elsewhere. But that’s an intangible benefit, and unattractive for that reason. That’s why you get told “The developers don’t read these forums”. That’s why Brainstorm acts mainly as a honeypot drawing noise away from the bug tracker. And that’s why it’s gradually getting more difficult to report a bug about Ubuntu. We *need* to erect those barriers, so that we have time left in the day to improve the software. This may seem outrageous to people used to contributing to smaller projects that have less noise — i.e. every open source project in the world, other than Firefox, Ubuntu, and possibly OpenOffice.org.
    (end quote)

    Note the “honeypot” part?

  13. I was fed up. I grabbed him by the shoulders, looked him in the eye and asked the question. Going through proper channels had no effect.

    All through my educational career, I have been frustrated by people not doing their jobs. In this case the guy failed to even consult with users what they needed and the bug-reporting system was horribly broken. They could easily have had bugzilla running at less cost.

    IT support at that place came around every few weeks, did stuff and left. They were on the road more than they did stuff. In fact, while we had the printing problems, they came to string cables to rooms where we did not have PCs… I made a suggestion to obtain Computers for Schools machines to put in those places but was shot down. I left after half a year. It was not my kind of place. I was quite capable of managing IT in the building for $0 which they well knew but I was kept “out of the loop”. It took me more than a week just to get a password for their system. It’s a good think Beast was along. I set up the students to boot PXE from Beast so we could get the job done. Imagine my courses having to wait weeks to get files set up on the server, or printing, etc. At the same time people were coming to my lab for help every five minutes while I was trying to do my job. No kidding. People had to install the right printer every time they wanted to print and it was buried deep.

  14. Linux Apostate says:

    “Finally, I physically assaulted an IT person and demanded to know why the problem had not been fixed.”

    This sounds very violent – did you really mean to write “assaulted”?

  15. Chris Weig wrote, “Giving feedback often results in “won’t fix””

    There are many reasons for “won’t fix” besides contempt for users. For example, the latest release deals with that, so why mess with the old code? Any user so in love with the old version can take the code, and change it as he wishes. There’s no need to feel put off. Much FLOSS evolves rapidly and distributes many new features with each release. One cannot expect a developer to maintain his previous work indefinitely any more than Ford maintains Model Ts. In fact, Ford does not maintain their products and farms that work out to dealers. The same can work for FLOSS. If you need something fixed without changing to a new release, hire a programmer.

  16. Clarence Moon wrote, “there is zero guarantee that anyone will be motivated to fix it. If it is fixed, there is nothing to ensure that the fix didn’t create more problems than it cured.”

    Hmmm. That sounds a lot like M$. See the EULA:
    “7” PRO…
    “A. LIMITED WARRANTY. If you follow the instructions and the software is properly licensed, the software will perform substantially as described in the Microsoft materials that you receive in or with the software.
    B. TERM OF WARRANTY; WARRANTY RECIPIENT; LENGTH OF ANY IMPLIED WARRANTIES.
    The limited warranty covers the software for one year after acquired by the first user. If you receive supplements, updates, or replacement software during that year, they will be covered for the remainder of the warranty or 30 days, whichever is longer. If the first user transfers the software, the remainder of the warranty will apply to the recipient.

    G. NO OTHER WARRANTIES. The limited warranty is the only direct warranty from Microsoft. Microsoft gives no other express warranties, guarantees or conditions. Where allowed by your local laws, Microsoft excludes implied warranties of merchantability, fitness for a particular purpose and non-infringement. If your local laws give you any implied warranties, guarantees or conditions, despite this exclusion, your remedies are described in the Remedy for Breach of Warranty clause above, to the extent permitted by your local laws.”

    That inspires confidence… You spend big bucks on your PC and after a year it can go POOF! and there’s no recourse for the software licence you paid. Of course, that other OS does not go POOF! It just slows down, picks up malware, messes up the file-system or refuses to boot.

    I trust Debian a lot farther than I would trust a convicted monopolist known to have corrupted competition globally.

  17. Chris Weig says:

    In FLOSS there are seldom well-defined means of giving developers feedback. Giving feedback often results in “won’t fix”.

    There. I fixed it for you, Mr. Pogson.

    FLOSS users are thought of as scum by most FLOSS developers. Life could be so nice for FLOSS developers, if only those pesky FLOSS users would go away.

  18. Clarence Moon says:

    So, if you want your feedback to count, if you want to use software that increasingly works for you and not someone else, use FLOSS, the right way to do IT.

    I say “Baloney!”, Mr. Pogson, baloney! We have almost 100 developers working on our product code and almost as many whose full time job it is to test new versions in process for new or revived defects. We have fully staffed, 24/7 available call centers for our customers to report problems and get help with solutions. If the problem is not fixed immediately by the call center tech, it gets a tracking number that subsequent calls can reference for progress and anticipated time for the problem to be treated. What does Debian have? Some email address, that’s all. Debian has no money and no paid staff to fix anything.

    If you can wend your way through obtaining the bug reporting tool and manage to log in the bug yourself with Debian, there is zero guarantee that anyone will be motivated to fix it. If it is fixed, there is nothing to ensure that the fix didn’t create more problems than it cured. Users are on their own to hope and pray that someone will smile on them and take care of their problem.

    Having access to source that you cannot understand is no substitute for having someone with some skin in the game having their feet held to the fire for a solution. There isn’t one person in a million who can even read such code and even fewer with any sort of insight into how to fix it. FLOSS is a myth in terms of responsive reaction to problems.

Leave a Reply