Silicon Investor (SI) -- The First Internet Community

STOCKTALK

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor. We ask that you disable ad blocking while on Silicon Investor in the best interests of our community. If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.

Technology Stocks : Information Architects (IARC): E-Commerce & EIP -- Ignore unavailable to you. Want to Upgrade?

To: sibe who wrote (3423)	11/20/1997 12:19:00 AM
From: tech	Read Replies (2) \| Respond to of 10786

>> RE: ERROR RATE << Is there some kind of industry average or standard? A while back the Gartner Group put out an est. of 2 to 4 thousand errors per 1 million lines converted. However, those were early numbers based on code converted manually. I believe that there is no set industry standard for the companies such as ALYD or others who have access to tools. The numbers should be dramatically less. In any case, even if you have 1 to 100 errors per 1 million lines, that will cause problems. The real issue here is what will be the source of these errors? I believe that most errors will be "human errors", i.e. the more dependent your tool is on having "bodies" to do conversions, the more prone you are to having errors. Wait to you see what happens when companies get back their code from the body shops in India who are using cheap labor to convert code, or the companies who are getting tools from PTUS who hope to do the conversion work internally. The simple equation is: The more your solution depends on bodies to convert code, the more errors you will have. Do you know the error rate for any Y2k company ? Yes, I know of one that expects an error rate of ZERO. ConSyGen (CSGI) has the industry's only fully-automated toolset. Their tool automatically finds and converts code. In Fact, NOT ONE LINE OF CODE IS CONVERTED BY A PROGRAMMER, the toolset does it all. A few months ago CSGI announced the first every fully automated conversion project that was successfully finished and tested. exchange2000.com

To: sibe who wrote (3423)	11/20/1997 12:49:00 AM
From: Jeffrey S. Mitchell	Read Replies (3) \| Respond to of 10786

Re: Y2K error rates Tech wrote: In any case, even if you have 1 to 100 errors per 1 million lines, that will cause problems... Yes, I know of one that expects* an error rate of ZERO.* Looking back at a previous post by "NJ Investors" regarding comments CEO Gruder made at the CR Conference about ALYD's error rate: Quality: 1 error per 100M LOC empirical* data; with expectation to be 1e/1 billion LOC. I should also point out that of all the programming errors ever made in the history of computing, NOT ONE SINGLE ERROR, Y2K or otherwise, has ever been attributed to FBN products. Better yet, FBN expects* an error rate of -100! That means we not only expect not to make errors, we actually expect to fix those made by non-FBN programmers. Hey, but what else did you expect? - Jeff