mission critical apps; was Re: cross platform ide

[nnoydb at excite.com]

Where might I locate the SDB/libIPC (to edcuate myself)?

K






 --- On Mon 02/09, Alex Rice < alex at mindlube.com > wrote:
From: Alex Rice [mailto: alex at mindlube.com]
To: use-revolution at lists.runrev.com
Date: Mon, 9 Feb 2004 11:41:13 -0700
Subject: Re: mission critical apps; was Re: cross platform ide

<br>On Feb 9, 2004, at 10:19 AM, Rob Cozens wrote:<br><br>> That is certainly not true of script syntax errors... and when you do <br>> encounter them you can open the script editor, correct the problem, <br>> and go right on with your runtime testing [unless applying the script <br>> changes the value of declared local variables].<br><br>I stand corrected. So some bugs are found when the script is edited. <br>Everything else is found at runtime.<br><br>><br>> That has just not been my general experience, Alex.<br><br>Did I say it *cannot be done*? No. I said there are unique challenges. <br>In a previous post you acknowledged as much:<br><br>> Upon reflection, I will acknowledge that message path issues were one <br>> of the most difficult parts of my transition from PL/1, C, & Pascal to <br>> HyperTalk.<br><br><br><br>>   And specifically, from a post to the rev_ipc group:<br>> As I write this, my OS X test SDB client has completed 24 hours of<br>> auto testing without interruption since I restarted it after turning<br>> off sleep mode on the iMac.  It has issued about 24,000 commands to<br>> the server since restarting.<br><br>Congratulations. I never claimed it was impossible to produce reliable <br>apps with runrev. Just challenging.<br><br>But I can get pedantic too. The messaging bug I was remembering that <br>Dar opened was this:<br><br>bug # 102: """After message ID of 2147483647 is assigned, the next <br>message ID is -2147483648.<br>  This in itself is not a problem; a large cycle in the message ID is <br>OK.  Both<br>pendingMessages and the result show a negative number. The problem is <br>that that messages with a negative message ID cannot be canceled."""<br><br>><br>> But I suggest that some of the issues being raised here cut to the <br>> very core foundation of X-Talks, and if you can't accept things the <br>> way they are, you will be continually frustrated with your experience <br>> with Revolution.<br><br>I sense you feel threatened by this thread because SDB and libIPC are <br>geared towards a more business like, maybe mission-critical <br>applications? For all I know SDB and libIPC are the most solid, <br>bulletproof apps ever engineered in an xtalk language. I wish you all <br>the best with SDB and libIPC.<br><br>--<br>Alex Rice | Mindlube Software | http://mindlube.com<br><br>_______________________________________________<br>use-revolution mailing list<br>use-revolution at lists.runrev.com<br>http://lists.runrev.com/mailman/listinfo/use-revolution<br>

_______________________________________________
Join Excite! - http://www.excite.com
The most personalized portal on the Web!