[SystemSafety] Collected stopgap measures

Peter Bernard Ladkin ladkin at causalis.com
Sun Nov 18 10:34:26 CET 2018


I find it odd that people are so ready to criticise Boehm and his work. It is forty years ago, and
it was pioneering. Of course there are some things we would do differently. There are some things
which Boehm himself did differently in his second large work, twenty years later. A dozen opinions
saying "Boehm didn't do it right" are unilluminating. Technical specifics, along with technical
criticism, are more helpful.

>> [Paul Sherwood] Just because Boehm's works are rather large, and widely cited, doesn't make them correct.

Size is not key. Being widely-cited is a decent pointer, provided that the citation count is not
because you made a mistake that everybody is keen to point out. The key point is that Boehm's work
has been extensively peer-reviewed in the reputable literature. That makes it quite likely to be
mostly correct.

> [Derek Jones] It shows the power of promoting one's work, which Boehm does heavily.

It shows the power of presenting one's work publicly and submitting it to peer review in the
highest-quality publications. As well as the power of, while not quite inventing the subject, taking
it forward substantially in a major work.

>> [Paul Sherwood] Laurent Bossavit went all the way down to the bottom a couple of the rabbit holes in [1], and from
>> that research Boehm was apparently just as prone to fallacious thinking as the rest of us.

As it stands, this is just "rumor" as Derek Jones would say. There seems to be a lot of it flying
around this list recently. Could I suggest you present Boehm's thesis, and Bossavit's refutation?

> [Derek Jones] He over-fits this data and quotes accuracy to two decimal places,
> while the data has a standard error at the level of one-digit
> accuracy.
> 
> There is also mistake in the fitting of one of his models:
> .....
> 
> I find this very surprising, given the data has been public for almost
> 40 years.  Why has nobody thought to check it and raise the issue?

They obviously have! You could write it up and submit it to IEEE Software Letters, or IEEE Trans.
Soft. Eng. for peer review.

FYI, I looked at the post. Easy to follow until I got to the word "mistake". You fail you say what
the mistake is. That is followed by a lot of what you call "rumor" or "anecdote": you say you did
this-and-that and didn't get the same results as this-and-that, but you don't actually present any
figures.

You couldn't publish this observation as written in a reputable journal, even as a letter. If your
observation is right, it is certainly worth publishing in such a forum. I encourage you to write it
up in a way in which people can follow your reasoning.

BTW, there is a collection of Boehm's papers in a book edited by Richard Selby, IEEE 2007.

https://books.google.de/books?id=ttaMIFv8bv8C&pg=PA174&lpg=PA174&dq=boehm+economics+Trans.+IEEE+Soft.+Eng.&source=bl&ots=yZklU1uRFp&sig=XSdVWanvjQkn0M3nkGgBTOlN5fo&hl=de&sa=X&ved=2ahUKEwifuan4wd3eAhWNuIsKHcL9D3g4ChDoATAAegQICBAB#v=onepage&q=boehm%20economics%20Trans.%20IEEE%20Soft.%20Eng.&f=false

PBL

Prof. Peter Bernard Ladkin, Bielefeld, Germany
MoreInCommon
Je suis Charlie
Tel+msg +49 (0)521 880 7319  www.rvs-bi.de





-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 833 bytes
Desc: OpenPGP digital signature
URL: <https://lists.techfak.uni-bielefeld.de/mailman/private/systemsafety/attachments/20181118/44d40d45/attachment-0001.sig>


More information about the systemsafety mailing list