Plateforme Level Extreme
Abonnement
Profil corporatif
Produits & Services
Support
Légal
English
Row buffering
Message
De
29/01/1999 16:57:03
 
Information générale
Forum:
Visual FoxPro
Catégorie:
Gestionnaire d'écran & Écrans
Titre:
Divers
Thread ID:
00180821
Message ID:
00182049
Vues:
12
Jim. . .

Firstly, we do disagree on the docs and their content. But let me make some comments on yours:

SNIP
>
>Jim,
>
>You and I disagree on the docs. I believe that they are fine, you think they are abysmal. Let's agree to disagree on this issue.
>
>It is unrealistic to think that every help entry will cover the entire possible scenarios of event sequences and give specific information for each possible sequence. The help states that you MUST set multilocks on in order to use CursorSetProp to set buffering. That is all ther eis to say on the matter. Through lack of comment one cna infer that set multilocks is not necessary when using a DE. A simple quick test proves this.

So, through lack of comment, I should infer that FLUSH *does* work on everyting, all the time?
Similarly I can infer that RELEASE will always release a variable and its assdociated memory all the time?
No, you know full well that that strategy is not productive.

>
>Since any document of the size of VFP's docs will have unaviodable errors in it, one should test things before depending on them anyway.
>
Yea, that could be true as regards errors, but not really so when it comes to omissions.
If it states that X will happen if I do Y, and X doesn't happen, I have a chance to do some digging and try to find out why.
But if it states that ISRLOCKED() (for a simple example) will report the status of the record for a lock but it doesn't tell me that this applies only if this app has placed the lock, how am I ever going to know?
How am I going to know that SET REFRESH works for more than BROWSE when the Help restricts it to BROWSE? And these are cases where something *is* said!

>The docs are not a training program, they are reference material and as such they should be brief and to the point. I don't want to read a forty page entry for Valid, I want brief information and if I need to know more I will investigate what I need to know.
>
I am not at all suggesting something like forty pages for Valid, and you should know that!
I am suggesting that design objectives which are relevant be communicated in the Help (and elsewhere, as they may see fit). This would give us a fighting chance to understand what is going on.
As a "reference", what good would a dictionary be to you if they arbitrarily left out words? The key to VFP is design objectives, and these *need* to be communicated to the users of the product.

>It appears that making the help system what you are asking for would render it useless to me. It would then be information overflow and confusing to read and understand. Each entry would require that every possible event sequence be discussed even if a sequence did not alter the behavior, because otherwise the docs would be "incomplete". The entry being discussed here is a perfect example of this. You say you want the docs to tell you that the stated requirement for cursorsetprop is not needed for a DE.
>
You know that is silliness (making it useless to you because of overload). And I don't know what I want it to say as regards Multilocks and CURSORSETPROP(), but I do know that it now leaves it as guesswork and that is unacceptable.

>Jim, if you want input to MS about VFP, apply for the betas. There you get alot of input. They even have a docs newsgroup for doc enhancement and correction issues. However, it does mean you have to provide some effort into testing the product for them.

I have done betas before (though never for MS) and I have had my fill, thank you. I now leave those for others going through the same phase I was going through at that time.
And Since I recollect Ms. Slater-Nicholl (sp?) stating that her attempts to get the Help on ISRLOCKED() fixed in VFP 3 beta fell on deaf ears, what possible hope would I have??

Jim, the Hacker's Guide (VFP 3 version), KB articles with answer of "working as designed", repetitive questions in forums like this and just plain attention to detail provide fertile ground for the VFP team to see what initially needs amending. They just don't do it. Yet they claim that Foundation Classes are to help the novice, so the feign care in promoting VFP to more people. The foundation classes are not integrated into the Help though (pick a method and try to find it in the index), so how they expect this to be so is beyond me.
And I can tell you that I have heard more than one person say that they dumped VFP because it doesn't woork as documented. Sure, you and me have lived through this, but newer folk just move on to something else.

Jim N
Précédent
Répondre
Fil
Voir

Click here to load this message in the networking platform