Results 1 to 16 of 16

Thread: Who gets questions answered and who doesn't?

  1. #1
    Join Date
    Jul 2008
    Location
    St. Louis, MO
    Posts
    10
    Thanked 1 Time in 1 Post
    Qt products
    Qt4
    Platforms
    Windows

    Default Who gets questions answered and who doesn't?

    I posted a question under the Qt Programming forum and have had no response from anyone over the past couple of days. I noticed that over this same period of time quite a few questions occurred after mine and many received responses, many receiving more than one response.

    What are the criteria for getting questions answered?

    I need to know this since I don't want to waste my time with the forum if the questions that I am posting are not going to be addressed in some way. I understand that this is a free site, but at the same time I need to determine if this forum is a resource that we can rely upon or not. If it is not then we need to seek help from other places.

    Sorry to be so blunt. I am trying to determine what are reasonable expectations. We are a small company and are under a shoe-string budget. If the forum is not a good resource then I want to know that up front so we can locate alternatives, maybe purchase support I don't know.

    Regards,

    George

  2. #2
    Join Date
    Mar 2008
    Location
    Houston, Texas, USA
    Posts
    277
    Thanks
    9
    Thanked 17 Times in 17 Posts
    Qt products
    Qt4
    Platforms
    MacOS X Unix/X11 Windows Maemo/MeeGo

    Default Re: Who gets questions answered and who doesn't?

    Yes this is a free site driven by people wanting to help others on their own free time. Yes this forum is reliable for getting help. Some people may not have an answer that you need. And other questions maybe easier to answer for that individual. Just be patient also I can say. Just don't expect a quick reply right then and there some times you might. While you're waiting for someone to reply or what not. Keep searching the forum for a possible suggestion that was previously asked and/or read the documentation more. Because you might have overlooked something.
    Qt-4.7.3 | Gentoo ~amd64 | KDE-4.7
    Aki IRC Client for KDE4 | Qt Documentation

  3. #3
    Join Date
    Jan 2006
    Location
    Germany
    Posts
    4,380
    Thanks
    19
    Thanked 1,005 Times in 913 Posts
    Qt products
    Qt4
    Platforms
    Unix/X11 Windows Symbian S60
    Wiki edits
    5

    Default Re: Who gets questions answered and who doesn't?

    Quote Originally Posted by gjshannon View Post
    I posted a question under the Qt Programming forum and have had no response from anyone over the past couple of days. I noticed that over this same period of time quite a few questions occurred after mine and many received responses, many receiving more than one response.

    What are the criteria for getting questions answered?
    Well, since the criteria and algorithm how questions are answered are top secret, I can't tell you. But with normal thinking there are following possible reasons why your question(s) are not answered:
    • You have asked the wrong way (no one understand the question or nobody has the time to read a whole roman to figure out what you want)
    • Maybe none of us knowing the answer


    I need to know this since I don't want to waste my time with the forum if the questions that I am posting are not going to be addressed in some way. I understand that this is a free site, but at the same time I need to determine if this forum is a resource that we can rely upon or not. If it is not then we need to seek help from other places.

    Sorry to be so blunt. I am trying to determine what are reasonable expectations. We are a small company and are under a shoe-string budget. If the forum is not a good resource then I want to know that up front so we can locate alternatives, maybe purchase support I don't know.
    Hey, we are doing this here just for our fun. Because we like Qt and we are willing to spend our time to help others! So why blaming and crying around! You are a company and we (mostly opensource developer who doesn't earn money) should do your job? No way! (And I don't understand what's a waste of time querying here at the forum? Write a question is about 3 minutes.)
    So I guess you have to alternatives: Changing our attitude or buy commercial support.


    Just as you don't have mentioned: You have got an answer!

  4. #4
    Join Date
    Jan 2006
    Location
    Warsaw, Poland
    Posts
    33,359
    Thanks
    3
    Thanked 5,015 Times in 4,792 Posts
    Qt products
    Qt3 Qt4 Qt5 Qt/Embedded
    Platforms
    Unix/X11 Windows Android Maemo/MeeGo
    Wiki edits
    10

    Default Re: Who gets questions answered and who doesn't?

    Quote Originally Posted by gjshannon View Post
    What are the criteria for getting questions answered?
    I can tell you how I find threads to answer...

    1. I tend to read threads that have 0 replies, so if you reply to yourself instead of editing your original post, I'm likely to miss it.
    2. I tend to stay away from threads and posts that have the word "urgent" in their content.
    3. I tend to stay away from threads that have meaningless titles (like "I need help" or "Question" or "It doesn't work")
    4. I tend to stay away from threads that are simply uninteresting because the same issue was already solved dozens of times and the thread author didn't bother to use the search engine (although this ruins our statistics so if nobody else answers, I usually post a solution or a link to it).
    5. I tend to stay away from threads where people forget to turn off their caps-lock key.
    6. I try to answer every thread that has not been answered even if I don't know the answer at the beginning - I open up assistant and try to find relevant documentation, that's enough most of the times.
    7. When you registered at QtCentre, you were sent an email containing tips about asking questions. Follow them, they are there for a reason.


    I need to know this since I don't want to waste my time with the forum if the questions that I am posting are not going to be addressed in some way.
    Maybe the question was simply missed? Try bumping up the thread, maybe you'll get an answer. Questions here are on average answered within an hour from the time they were posted.

    I understand that this is a free site, but at the same time I need to determine if this forum is a resource that we can rely upon or not. If it is not then we need to seek help from other places.
    If the question is extremely complex, you might not get an answer - neither here not anywhere else. Live with it, my questions almost always remain unanswered

    Sorry to be so blunt. I am trying to determine what are reasonable expectations. We are a small company and are under a shoe-string budget. If the forum is not a good resource then I want to know that up front so we can locate alternatives, maybe purchase support I don't know.
    Being blunt is one thing, but being rude is another. If you discourage people, they might not answer you even if they do know the answer. Quoting Axel's motto - "It's nice to be important but it's also important to be nice".

    Edit:
    I just made a search on your posts. Correct me if I'm wrong (I know I'm not so that's just an expression...) but you have posted one question and you posted it yesterday afternoon (European time) possibly when most developers already left their work places to go home for the weekend and forget about programming and solving Qt issues and now after 24 hours during the weekend you complain about your single and only question not being answered? Come on...

    I could have answered your question but I can't possibly answer all the questions, I want others to contribute as well. Otherwise the site would be called "Wysota Centre" not "Qt Centre"...
    Last edited by wysota; 14th March 2009 at 19:37.
    Your biological and technological distinctiveness will be added to our own. Resistance is futile.

    Please ask Qt related questions on the forum and not using private messages or visitor messages.


  5. #5
    Join Date
    Jan 2006
    Posts
    132
    Thanked 16 Times in 16 Posts
    Qt products
    Qt4
    Platforms
    Windows

    Default Re: Who gets questions answered and who doesn't?

    If getting support in defined time frames is critical for your development, I suggest you buy a commercial Qt license (so have I, because I'm doing SW development for business).

    Free resources such as this forum or the Qt mailing lists are great, but there is no guarantee that your questions get answered at all, or to your satisfaction. As others have pointed out, it helps however very much to

    1. Read up documentation and search the forum or google. Nobody likes to answer that same questions over and over.
    2. Reduce your problem to a small as possible testcase, describe in simple and clear way, and provide a reasonable amount of side information. A small sample is the best way to get helping of ground. (Well this is also true when you request help from the formal Qt support, not only from free resources)

  6. #6
    Join Date
    Jul 2008
    Location
    St. Louis, MO
    Posts
    10
    Thanked 1 Time in 1 Post
    Qt products
    Qt4
    Platforms
    Windows

    Default Re: Who gets questions answered and who doesn't?

    Sorry for being devious.

    I was purposefully somewhat rude and aggressive.

    I had to test the waters to see what was necessary for a FAST response.

    We are dirt poor and have no cash for a support agreement. I've been out of work for a long time and we have one small contract.

    I need to determine the viability of Qt for a small business like ours who is bootstrapping.

    I may have to migrate to Java since the support/documentation for that seems broader than available for Qt.

    I prefer not to, but unless we can dig up the modest cash we don't have a lot of choice.

    Mr.Lykurg - don't take it personal! We are not a big cash cow company taking advantage of you! We are struggling to make ends meet and we don't have the luxury of a support agreement.

    Sorry to offend everyone. I have to make business decisions and unfortunately my creditors aren't very nice to me.

  7. #7
    Join Date
    Jan 2006
    Location
    Warsaw, Poland
    Posts
    33,359
    Thanks
    3
    Thanked 5,015 Times in 4,792 Posts
    Qt products
    Qt3 Qt4 Qt5 Qt/Embedded
    Platforms
    Unix/X11 Windows Android Maemo/MeeGo
    Wiki edits
    10

    Default Re: Who gets questions answered and who doesn't?

    Quote Originally Posted by gjshannon View Post
    I was purposefully somewhat rude and aggressive.
    Then don't be.

    I had to test the waters to see what was necessary for a FAST response.
    The test may backfire, remember that. There is nothing that will guarantee a fast response but there are things that may slow it down, so try to avoid them.
    Your biological and technological distinctiveness will be added to our own. Resistance is futile.

    Please ask Qt related questions on the forum and not using private messages or visitor messages.


  8. #8
    Join Date
    Jan 2009
    Posts
    78
    Thanks
    21
    Thanked 1 Time in 1 Post

    Default Re: Who gets questions answered and who doesn't?

    Well ... i've been making lot's of newbie questions and always getting an answer.
    Just want to thank people who spend their time helping others and expecting nothing.
    THANKS

  9. #9
    Join Date
    Sep 2007
    Location
    Rome, GA
    Posts
    199
    Thanks
    14
    Thanked 41 Times in 35 Posts
    Qt products
    Qt4
    Platforms
    MacOS X Windows

    Default Re: Who gets questions answered and who doesn't?

    What a way to introduce yourself to a forum. Jeez. I can understand your concern, but you can clearly see from the views/replies that questions generally get answered here. For a comparison see qtforum.org. As far as I know this is the best place for Qt-related questions.

    PS - Welcome!

  10. #10
    Join Date
    Jul 2008
    Location
    St. Louis, MO
    Posts
    10
    Thanked 1 Time in 1 Post
    Qt products
    Qt4
    Platforms
    Windows

    Default Re: Who gets questions answered and who doesn't?

    Hey Wysota:

    By the way, I offered to use of our semantic search engine to Nokia for Qt use FOR FREE. I'd like to see if we can make it a lot easier to ask newbie questions, especially for the stuff that stumps a newbie but is second nature to gurus.

    The objective is to leverage an ontology of Qt documentation concepts via our semantic engine so that it is possible to get answers to simple questions really FAST.

    That way us newbies won't have to hit the forums (or buy commercial support) for the simple stuff.

    I have also offered my support to generate improvements to Qt documentation, pending release of the contribution model to see if we can participate.

    Just so you know that I am not the complete heartless slob you probably think I am (I'm actually not as bad as it seems, just have the thankless job of having to push some buttons once and a while).

  11. #11
    Join Date
    Jan 2006
    Location
    Warsaw, Poland
    Posts
    33,359
    Thanks
    3
    Thanked 5,015 Times in 4,792 Posts
    Qt products
    Qt3 Qt4 Qt5 Qt/Embedded
    Platforms
    Unix/X11 Windows Android Maemo/MeeGo
    Wiki edits
    10

    Default Re: Who gets questions answered and who doesn't?

    Quote Originally Posted by gjshannon View Post
    Hey Wysota:

    By the way, I offered to use of our semantic search engine to Nokia for Qt use FOR FREE. I'd like to see if we can make it a lot easier to ask newbie questions, especially for the stuff that stumps a newbie but is second nature to gurus.
    I wanted to implement an NLP engine in a form of a bot for Qt-related content myself but I didn't find time to do it.

    The objective is to leverage an ontology of Qt documentation concepts via our semantic engine so that it is possible to get answers to simple questions really FAST.
    Honestly, to get simple questions answered you don't need extra semantic measures. Why? First of all google already uses a semi-semantic approach, so searching through google will let you use synonyms and misspelled words. Second of all Qt docs are written in a very simple language so if you ask the search engine for simple terms, it is likely to return good results. Third of all many people here have very poor english skills and I'm guessing your engine is tailored for the english language so it will probably have trouble tolerating polish, indian or chineese syntax and homonyms in english sentences.

    That way us newbies won't have to hit the forums (or buy commercial support) for the simple stuff.
    Nokia earns money on support so I wouldn't expect them to be very happy about it

    Just so you know that I am not the complete heartless slob you probably think I am (I'm actually not as bad as it seems, just have the thankless job of having to push some buttons once and a while).
    I'm not judging people like that. If I were, I would have to judge myself this way too and I could not be very happy about the outcome of such judgegement There is nothing wrong in being irritated although I admit your level of patience seemed quite low. Unfortunately you have to live with the fact that we have our own jobs and nobody pays us to sit here and solve people's problems.
    Your biological and technological distinctiveness will be added to our own. Resistance is futile.

    Please ask Qt related questions on the forum and not using private messages or visitor messages.


  12. #12
    Join Date
    Jul 2008
    Location
    St. Louis, MO
    Posts
    10
    Thanked 1 Time in 1 Post
    Qt products
    Qt4
    Platforms
    Windows

    Default Re: Who gets questions answered and who doesn't?

    Google is not as good as folks think.

    First of all the the use of keywords for the user interface. While it may (or may not be) be semi-semantic, the user interface is a keyword-based technology which is quite a poor substitute for phrases that have a meaning.

    In the healthcare arena our engine was tested by a physician and shown to be 290% more accurate than keywords.

    To get this we use a medically approved ontology of terms, consisting of 300K terms and over 1.3 million relationships.

    PLUS, it works with absolutely no keywords. Enter a description of a problem or question in your own language and hit the search button and you are done.

    For example, how accurate are keywords when searching for "Dorsolumbar spinal fusion using Harrington rod"? In Google that is a mess of keywords and clauses to search. Our engine uses a set of robuts topology equations for comparing ontology covering spaces instead. Absolutely no keywords.

    In our search engine that is ONE concept, not multiple keywords. Google apparently is "semi-semantic" with an advertised set of rules that number about 100 in size. Hence we get 290% greater accuracy with ONE equation and no rules.

    Furthermore, Google relies quite heavily upon click behavior of users to apply their supposed semantic search (this information I received directly from a VP level person at Google). So they added a heuristic that weights results basic upon the fact that a keyword search may look for "diabetes" but the user clicked on something different. This is a heuristic that pushes the behavior of the masses on one search.

    If a physician wants information on "Dorsolumbar spinal fusion for Harrington rod for patient with history of post-poliomyelitis, lumbar pain, and stooped posture" then we can automatically extract all medical concepts and run a very accurate search. Google chokes on this.

    So, my idea is to do two things:

    a) enable very complex searches using phrases to describe a problem or question ("No right-click mouse event processing for a scene when creating scene items that consist of subclass of QObject and QGraphicsItemGroup with reimplementing the paint method"). This is much closer to what users THINK than just one or two keywords.

    b) respond to a search with the ability to both get accurate information but also provide the user with a learning path to understand the results. For example, it may be required that the user understand the basics of how QGraphicsItemGroup processes events and computes QRectF before they can understand how it interacts with events in the scene (that may be a dumb example, but you get the idea). It is an application of basic learning theory.

    To do this the open-source community would need to commit to creating an ontology for Qt that is used by the search engine to provide a very easy interface for complex queries with accurate results. Such an ontology is far smaller than the ontology I mentioned for medicine.

    Nokia is using Qt as a PURE STRATEGY PLAY. That is, no revenue only a way to push Nokia sales. I've been in the IT business for over 20 years, working on the management team and reporting to the CEO of an IT business, and I can almost guarentee that the last thing that is good for Nokia is an influx of low-margin LGPL maintenance contracts that requires that they hire more staff. These maintenance contracts are notorious for small margin profits. Doing this significantly increases their risk.

    But what they definitely want is rapid adoption of Qt to push Nokia devices. Hence anyway to get a user up quickly and easily brings them that much closer to achieving that goal.

  13. #13
    Join Date
    Jan 2006
    Location
    Warsaw, Poland
    Posts
    33,359
    Thanks
    3
    Thanked 5,015 Times in 4,792 Posts
    Qt products
    Qt3 Qt4 Qt5 Qt/Embedded
    Platforms
    Unix/X11 Windows Android Maemo/MeeGo
    Wiki edits
    10

    Default Re: Who gets questions answered and who doesn't?

    Quote Originally Posted by gjshannon View Post
    First of all the the use of keywords for the user interface. While it may (or may not be) be semi-semantic, the user interface is a keyword-based technology which is quite a poor substitute for phrases that have a meaning.
    I didn't say it's a replacement for a semantic web. I said you won't get many benefits in this particular case - when searching for Qt knowledge.

    In the healthcare arena our engine was tested by a physician and shown to be 290% more accurate than keywords.

    To get this we use a medically approved ontology of terms, consisting of 300K terms and over 1.3 million relationships.
    That's because this is a wider topic. With Qt there are practically two sources of knowledge - one is the reference manual, the other is community sites (I'm counting Qt Quarterly as part of the "reference" here). On the community sites the knowledge is in reality extracted from the reference manual as well only that it might be passed using different words (synonyms). You won't get hidden relations that might be deduced or induced simply because the relations are very clear and the language used is very simple so that more people can understand it. So to summarize - we are dealing with a single "knowledge source/domain" here. If you wanted to reference this knowledge to some other knowledge (like about user interfaces in general) then semantic approach would have its benefits but with the current situation my guess is that it won't get better results than using keywords, especially that the context of the sentence is clearly indicated by the keywords themselves (like when using class or component names).

    PLUS, it works with absolutely no keywords. Enter a description of a problem or question in your own language and hit the search button and you are done.
    You see the problem is that I (a human) have sometimes trouble understanding another human even if she/he uses the same (english) language. If that's the case, I'm not very confident we could write rules to make machines extract and understand knowledge contained in such posts. You can try teaching your web based on the posts here, I can give you some user names whose posts you can use as an exam set for the network

    For example, how accurate are keywords when searching for "Dorsolumbar spinal fusion using Harrington rod"? In Google that is a mess of keywords and clauses to search. Our engine uses a set of robuts topology equations for comparing ontology covering spaces instead. Absolutely no keywords.
    That's fine but it won't work with some of the posts here Keywords won't help (or mess) either, trust me.

    In our search engine that is ONE concept, not multiple keywords. Google apparently is "semi-semantic" with an advertised set of rules that number about 100 in size. Hence we get 290% greater accuracy with ONE equation and no rules.
    Hmm... AFAIK this can be a single keyword in Google as well. Just remember Google is not tailored to a specific domain. Although in health care domain this expression might mean one concept, there are domains where it can mean many different things. I guess if we fed your engine with "Dorso liumpair backbonal combining use stick from harringtoun" it might have trouble determining that you meant the earlier mentioned concept.

    Furthermore, Google relies quite heavily upon click behavior of users to apply their supposed semantic search (this information I received directly from a VP level person at Google). So they added a heuristic that weights results basic upon the fact that a keyword search may look for "diabetes" but the user clicked on something different. This is a heuristic that pushes the behavior of the masses on one search.
    Trust me, I'm not defending Google But they are the best and most widespread search engine around so apparently they have the best results on average. I'm sure this situation will change at some point and even Google will yield to a semantic approach but probably that won't happen tomorrow.

    a) enable very complex searches using phrases to describe a problem or question ("No right-click mouse event processing for a scene when creating scene items that consist of subclass of QObject and QGraphicsItemGroup with reimplementing the paint method"). This is much closer to what users THINK than just one or two keywords.
    Unfortunately practice shows that this problem would be described as "right-click doesn't work with paint" or even as "mouse click not making".

    You know, it all looks pretty and simple for a person that has been dealing with American English for the whole life Unfortunately (although personally I don't mind) still more people speak Chinese than US English.

    To do this the open-source community would need to commit to creating an ontology for Qt that is used by the search engine to provide a very easy interface for complex queries with accurate results. Such an ontology is far smaller than the ontology I mentioned for medicine.
    Ok, I'll be damn straight here - I don't think this will happen. We can't make people use "[code]" tags in their posts, I don't think anyone is able to make them create an ontology for Qt.

    The bright side is that my doctoral research and (hopefully) thesis is oriented around an ontology for describing user-interfaces and using it for UI validation and verification, so not all is lost.

    I can tell you what was my idea for an approach (which was of course very simple). First of all I wanted to index the reference, QQ, maybe also the posts here and have some statistics of combinations of words that go together. Then after someone posted a question, we would correct most common spelling mistakes, prune stopwords and look for similar senteces in the indexed content based on those statistics and maybe some heuristics (based on class names and keywords probably) as well. Knowing that answers to most problems can be found directly in the docs, we would cover a quite large area this way. The bot would answer by quoting a sentence that it thinks answers the question. So the pseudo-semantics would be hard-coded here in those heuristic rules and statistics. We would ignore grammar problems and spelling problems as these can't be overcome in our domain. You could call this approach silly but in my opinion it could give accurate answers in about 30-40% of the cases. It should work better than Google, that's certain.
    Your biological and technological distinctiveness will be added to our own. Resistance is futile.

    Please ask Qt related questions on the forum and not using private messages or visitor messages.


  14. #14
    Join Date
    Jul 2008
    Location
    St. Louis, MO
    Posts
    10
    Thanked 1 Time in 1 Post
    Qt products
    Qt4
    Platforms
    Windows

    Default Re: Who gets questions answered and who doesn't?

    I don't want to get into a point for point rebuttal, but in general it is not necessary to tag anything with my engine. The point is that a small group would create the ontology and the users literally do nothing more than describe what they want.

    It appears that I need to create a prototype to prove its abilities.

    By the way, a very similar search engine came about in parallel with ours from a university in florida that had a contract with the US Space Systems Command. We were unaware that they were doing similar work in parallel with us. They published the results in a book and blew the doors off of key word search, including generating a LOT of test results to prove it.

    FYI, I have been in contact with someone doing work on property models with ontologies for user components. His doctoral professor helped develop the property model components used by Adobe. He is very positive that significant new and more user-friendly components will come from this.

  15. #15
    Join Date
    Jan 2006
    Location
    Warsaw, Poland
    Posts
    33,359
    Thanks
    3
    Thanked 5,015 Times in 4,792 Posts
    Qt products
    Qt3 Qt4 Qt5 Qt/Embedded
    Platforms
    Unix/X11 Windows Android Maemo/MeeGo
    Wiki edits
    10

    Default Re: Who gets questions answered and who doesn't?

    Quote Originally Posted by gjshannon View Post
    I don't want to get into a point for point rebuttal, but in general it is not necessary to tag anything with my engine. The point is that a small group would create the ontology and the users literally do nothing more than describe what they want.
    Yes, I understand that but my point is you won't find that small group.

    It appears that I need to create a prototype to prove its abilities.
    You don't have to prove anything, I'm completely aware of the possibilities here. I just say in this particular case it might not be worth the effort. Although of course please do prove me wrong, if you have some free time to spare.

    FYI, I have been in contact with someone doing work on property models with ontologies for user components. His doctoral professor helped develop the property model components used by Adobe. He is very positive that significant new and more user-friendly components will come from this.
    Is the work available somewhere?
    Your biological and technological distinctiveness will be added to our own. Resistance is futile.

    Please ask Qt related questions on the forum and not using private messages or visitor messages.


  16. #16
    Join Date
    Mar 2009
    Location
    California, USA
    Posts
    31
    Thanks
    2
    Thanked 7 Times in 6 Posts
    Qt products
    Qt3 Qt4
    Platforms
    Unix/X11 Windows

    Default I love this forum!!

    My first program compiled and ran in November of 1963. God, I am sooo old.

    I am just learning Qt and I am very impressed with this forum and the work of the people who answer questions on this site. My last language was Java and I have spent a lot of time on Sun's Java Forums. On those forums, people are very rude and disrespectful, particularly of newbies.

    Oh, and I am very impressed with Qt Creator and the Qt implementation of C++. I know I will ask some dumb questions too.



    Gary
    Last edited by Chisum; 19th March 2009 at 22:04. Reason: spelling error

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Digia, Qt and their respective logos are trademarks of Digia Plc in Finland and/or other countries worldwide.