TechSpoken

"Any ideas?" is the most frequently-asked question in technical forums. My answer is: yes.

I gotta ask again... I gotta crow... and you gotta voice and a vote

Work is still horrendously busy, so I'm slow getting here in general.  Heck, I'm slow getting my laundry done.

But this weekend I have promised a number of people on the MSDN RS forums a fairly comprehensive and difficult-to-write walkthrough, so I'm hoping to get through that.  I'm gradually writing, there are just a lot of different pieces to that particular puzzle, and it looks like I will be posting it in two parts in the next couple of days.

Meanwhile... I was hoping I was going to get some comments/requests/opinions when I posted http://spacefold.com/lisa/post/2008/02/03/Beyond-TMM-analysis-of-a-bug-and-aftermath-of-architecture-work-What-happens-now.aspx from the VFP folks about a week ago.  High ratings are nice, thank you folks, but nobody wants to weigh in or vote on the plan they think is best?

Also... I gotta take a couple of seconds out here to say: my kid is really something. Please read http://googlepublicpolicy.blogspot.com/2008/02/rep-markeys-net-neutrality-legislation.html.  You can take this glowing reference with a grain of salt; Derek certainly does. 

Inside the tech world or outside: Please stand up and be counted.

 

 

Comments (6) -

  • Carl Warner

    2/21/2008 2:30:36 AM |

    I really do appreciate your writing flourishes.  But, I have pointed others to your posts and their reply to me was "what is she talking about?"  I hate to say this, but I think a dumbed down direct post might be in order.  They are the same ones who get to vote in November.  Just saying...

  • L

    2/21/2008 4:05:53 AM |

    Yeah, well.  Thanks for the feedback, Carl.

    I am what I am. I sometimes write walkthroughs meant to provide soup-to-nuts directions for something...  and other times I write how I feel like writing.  

    I realize that both approaches are not equally appreciated, but I can't do otherwise.  This is especially true when I have some very strong feelings to tamp down to get myself to write on a subject at all -- which happens to be the case here.

    <i>It's my blog.</i> If I can't write the way I want to write here... I should just stop writing.  I'm close to that anyway.

    I do appreciate informing me that, as is too often the case, I have not done what I intended to do by writing at all.

    <b>To the community at large</b>

    If nobody has a better idea, or is strongly opposed, at the current time it looks like the _REPORT* apps source is going up on CodePlex, with Bo Durban as project manager.

    Please speak up if you have something to say.

    How's that?

  • Cesar Chalom

    2/26/2008 1:13:17 AM |

    Hey Lisa,

    The new report features and possibilities are really terrific. But IMO, the vast majority of the VFP developers will start using them only when they will need one of those new features, and at the same time find some samples to start working with.

    In my personal case, I use the report listeners for some specific tasks, and for my case, they work very well. I've faced the DataGroup bug that was reported by Cathy Pountney here:
    cathypountney.blogspot.com/.../...g-with-data.html
    cathypountney.blogspot.com/.../...a-group-bug.html
    cathypountney.blogspot.com/.../...a-group-bug.html

    But there are workarounds for it, and I'm living quite well with SP2.

    But many other people are still reluctant in "upgrading" to SP2 due to some bugs.

    Are those bugs fixable ?

    Please understand that most people are really beginners on all this new world of report listeners. Few people tried to go deeper on this till now.

    I'm one of those that gave a rate on that post. I think it was the best way to say to you: "THANK YOU FOR YOUR EFFORTS AND YOUR COMMITMENT IN SOLVING THESE ISSUES - I APPRECIATE THAT, but I don't feel I have skills to send you more comments on this, and I will support your decision." This is what my 5 star rating meant to be.

    Adding this project to Codeplex will help some people to get involved, but I antecipate that you'll be able to count the amount of these people in just one hand, unfortunately.

  • L

    2/26/2008 2:51:24 AM |

    Hi Cesar,

    Thank you for your comments.  Actually there are several comments here -- I will try to answer each one of them if I can, or as best I can...

    1) I am glad that you have a workaround for whatever bug Cathy is talking about, and I hope you share it with other people.  Knowing you, I am confident that you do.

    That is all any of us can actually ask of any others of us.  Share what you know, to the limit that you can.

    2) Colin and I try to share samples and information as best we can, like anybody else.  That is all you can ask of us.

    I have completed the core topics in TMM, and there are some samples included there. I expect there will be more, but I have *no* time to work on this right now.  (We have just moved and started new jobs.)

    We have published Colin's DataViz articles, and they have an end-to-end, fully-worked sample that you can follow step by step in simple iterations.

    By fully-worked I mean that this sample touches on all the basics of reporting extension, both design time and runtime.  I should also tell you that both Colin and I fully-worked very on that sample, the progression of the instruction throughout the article, and the lucidity of the presentation.

    3) You ask "are those bugs fixable".  I have to answer "It depends on which bugs"!  

    IOW: I can't answer this in any sort of 100% blanket way.  

    Right now, I can tell you only that fixing the Xbase bugs will have the best chance of getting done -- whether only a handful of people participate or more than a handful -- if the project is accessible on Codeplex, which is why it is going there.  So, it is a start.

    I can also remind you that the Xbase portion of the reporting improvements in SP2 are usable in SP1. As a result, even though I don't know when and how any SP2 product bugs might be fixed, I don't believe that should stop you from using the new stuff we released for reporting.  

    Look: There are only a few base product changes involved in SP2 underlying the reporting improvements, and none of them are required by the Xbase changes.  The Xbase pieces will work in SP1.  

    As an example, there is a LoadReport improvement in the base product. The Xbase piece that leverages it, gfxNoRender, figures out how to behave when it is running in SP1 and adjusts.  

    It is true that the few base product changes in SP2 that were explicitly made in reporting were, I believe, accompanied by some bug fixes to account for previously reported bugs.  I was not involved in that process and can't say much about it.  

    Perhaps one of those bug fixes caused whatever issue Cathy is talking about. I couldn't say, and I have no information to give you about whether it will be fixed. I also have no opinion on whether or not it should stop people from upgrading, whether it is serious, etc.

    All I can do is tell you that any changes from SP1 to SP2 should not stop people from using the improvements in SP2 reporting -- because 95% of them were Xbase.  

    Is that 95% of the changes, which is Xbase, perfect, no bugs? I am sure it is not.  Hence this blog post. Is it fixable? I am sure it is.  Hence Codeplex enters the conversation.

    4) I do take your point that nobody will use this stuff until they have a need -- but that is like anything else, isn't it?  

    * -- If they are not using it because they don't have a need, it will be there for them when they do.  

    * -- If they have a need but are not using it because they are afraid of it, or because somebody is making them afraid of it... that would be a shame, in my opinion.

  • Cesar Chalom

    2/26/2008 6:02:51 AM |

    Hi Lisa,

    1 - About bugs on the DataGroups
    Before Cathy or Bo published their solutions, I just came back to the solution I used in previous versions of VFP, adapting the data... Nothing elegant, but worked.


    2 - About sharing samples
    Till the middle of last year, all we had were 2 couples of articles, from Doug Hennig and Cathy Pountney, and some few report listeners published in www.reportlistener.com , that unfortunately is not available any more.

    I truly understand how busy you've been last year, creating the enhancements for SP2. Fortunately, you and Colin started publishing some of the materials that you've presented in some Conferences / User Groups. Those 2 introductory articles from Colin have already saved me many days.

    And I know that the situation we have at this moment is much different, having the HELP files, and the articles from you and Colin.


    3 - About "You ask "are those bugs fixable". I have to answer "It depends on which bugs"!"

    Sorry, I was not clear on that question. I'd like to know if the DataGroups bug. I don't know if it needs to SP2 core to be updated or just a fix in the Report*.app files would be enough.


    4 - About people not upgrading to SP2
    Unfortunately, this is happening. People are still waiting for SP2a or a patch, whatever. I'm using SP2 and I'm happy with it, but this is not what the majority say.


    5 - You said: "If they have a need but are not using it because they are afraid of it, or because somebody is making them afraid of it... that would be a shame, in my opinion."

    I think that definitely it is not the case here.
    In my opinion, it is about the lack of skills.

    Last year I published in my blog a ReportListener sample weblogs.foxite.com/.../3641.aspx  that allows to have full justified texts in reports (at this moment, we have only the possibility for Center, Left or right alignments). IMO, the published sample was so easy to reproduce, but I received lots of emails complaining that the sample did not work...

    But the problem was that most of them could not even use a simple class... I had to teach them some basic things. :-(

    The things that people use more often are those that are accessible directly via the report designer, like the super cool "label rotating" and the "dynamic formatting".

    So, people will need some more time, till they get used to use the new features, and till more people publish new samples. Maybe I'm wrong, but I think that most people first want to see the results, a cool output, and only after this, they try to understand how it works.

    Thanks for all this great job you and Colin have been doing. It's just a matter of time till most people will get used to work with them.

  • L

    3/1/2008 4:09:49 AM |

    Cesar, although I thank you for your comments, I found the job of answering them too depressing and avoided it for a week.  I do have the excuse of a 16-hour work day, but what the heck.

    ::
    Till the middle of last year, all we had were 2 couples of articles, from Doug Hennig and Cathy Pountney, and some few report listeners published in www.reportlistener.com , that unfortunately is not available any more.
    ..
    Fortunately, you and Colin started publishing some of the materials that you've presented in some Conferences / User Groups. Those 2 introductory articles from Colin have already saved me many days.
    ::

    You know what: The original helpfiles, with RTM, had the basis of this architecture described in it. With fully-worked examples. What we did in SP2 was implement it in-the-box.

    The examples were in two help topics. Prior to SP2, I constantly referred people to these help topics.  All they had to do was try out that code.  

    Doug's article, in fact, on effects listeners came directly from that code or from a series of recommendations I made to him in the beta cycle, which I made to him as I was *writing* those help topics (I still have the beta cycle messages, if anybody is curious).

    **There were two help topics** (not a great burden to ask anybody to read** and they **fully worked out** how you did this stuff on the runtime side.

    We expected the design-time side to take some vendors to implement, if you wanted a nice interface for it, and we provided help to various people for that. There were additional helpfile topics we could point them to, we didn't hide anything.  But the two sides are not coupled and you can do the runtime work pretty easily if you followed those two help topics.

    If you look at the "Report XML MemberData Extensions" article,
    you will see part of it, in either version of the help topic.  If you are looking in the SP2 version of the help topic, you will see that it points you to "Considerations for Creating New Report Output Types", but you will not see the code that "Report XML MemberData Extensions" points to in "Considerations for Creating Report Output Types" - SP2.

    And believe me I was very disheartened when I saw that.  We did expand and improve the whole FX thing in SP2 so, as implemented in-the-box, you wouldn't want to follow "Report XML MemberData Extensions" exactly. They should have pulled the code from both, and added pointers to appropriate FFC topics, or left them both in and just explained that these examples were workable as-is, when paired, but that FFC information was now better in SP2.

    Oh wait.. I forgot.  The SP2 helpfile didn't even cover the FX and GFX additions.  Well, now you know why http://spacefold.com/articles/TMM/ was important to me, and why the topics covered so far are the ones I considered essential.

    Never mind.  You had more help than you say. We didn't stop giving when we ran out of budget or spec or anything else in the SP1 cycle. We stopped when we ran out of time on the release, which was (believe me) way after we ran out of budget and spec.  

    So Carl says "well, people don't like the way you explain it". And you say "well, people need more help". And I have to say "this is the limit of what I have to give you".  And there we are.

Comments are closed