[Shootout-list] Fasta in OCaml

[William D. Neumann]

On Dec 20, 2004, at 3:15 PM, Isaac Gouy wrote:

> William, is it too much to ask that when you notice the "sample output
> file doesn't jive with the benchmark program" you post that information
> to the list, before you waste any of your valuable time?

Unfortunately, this time I assumed the output shown in the Oberon page 
was correct (and that the Oberon program was also correct), and 
validated my code against that before I ever looked at the sample 
output file.  In the future I will check to make sure the output of the 
"correct" programs matches the sample output file before I start 
coding.

And apparently, it wouldn't have mattered too much if I had looked 
first, as sometime between last night and this morning, when I 
submitted thefirst  OCaml version, the page was regenerated with the 
current Lua program replacing the old one (that, I assume, worked like 
the Oberon and Chicken submissions), and the Python submission added -- 
although the test validation file must not have been updated then.

Now I don't know if the sample output file agreed with the 
Oberon/Chicken/old Lua programs last night (I didn't look at it then), 
but whether it did or not, it brings up another question: Why was the 
benchmark changed *after* promotion to the front page?  I can 
understand changing them before they move from the sandbox to the main 
page (as fannkuch, etc. did which is why Brent needed to alter the 
output part of my submissions a bit), but after it hits the front page? 
  Without so much as an announcement to the mailing list?  What's up 
with that? (And as another point, wouldn't it be good to announce to 
the list when a benchmark is added or removed?)

> Are you offering to spend your time keeping the website consistent?

Well, I thought that things were more or less automatically generated, 
and I would have assumed that the sample output file and the test 
validation file would have been synched up automatically.  But a: had I 
known the benchmark had changed overnight, and b: had access to the 
site's server, yes, I would have fixed it this morning when I noticed 
it was in an inconsistent state (Of course, had I known the benchmark 
was changing in a way that I knew would break some of the submissions, 
I probably would have checked that a: those submissions did break, and 
b: the newly working programs actually did work).

William D. Neumann

"You've got Rita Marlowe in the palm of your hand."
"Palm of my hand?  You haven't seen Rita Marlowe..."

		-- Will Success Spoil Rock Hunter?