RE: [abc] oopsla06abc

On Mon, 6 Mar 2006, Eric Bodden wrote:

> Hi all.
>
> I would just like to know if I could already start to fill in some of
> the gaps in the OOPSLA paper even before indexing and number taking is
> finished. I thought that probably Ondrej and me could already have a
> look at the future work section, incorporating ITD optimizations, and
> static approximation, since I think that's not gonna change any more. I
> also think I can already give the rough idea about what a static
> approximation should look like, since I have started on the
> implementation for Laurie's compiler course.

I was thinking we want to let that future work section be driven by the
numbers. So for example, we already know that ajhotdraw is not going
to benefit by indexing; and we're a lot slower than the aspect there.
What analyses might improve that situation? [for this example, the scheme
for compiling to itds described in aosd05tmimpl should do it]
We'll need to see the whole table of numbers before we know for sure
what other examples to address in this section.

You could draft a few paragraphs that need to go in there anyway,
for instance reviewing all the previous work that does the safeenum
example statically.

> I can also try to polish sections 2, 3.1. and 3.2 but maybe that's still
> a bit early right now. Sections 4 and 6 will need time until we have
> final numbers, I reckon.
>
> Eric
>
> P.S. J-LO should be finished by tonight ;-)
>
>> -----Original Message-----
>> From: Majordomo list server
>> [mailto:majordomo@comlab.ox.ac.uk] On Behalf Of Oege de Moor
>> Sent: Friday, March 03, 2006 5:27 PM
>> To: abc@comlab.ox.ac.uk
>> Subject: [abc] oopsla06abc
>>
>>
>> I've created a new cvs module for this paper on musketeer.
>>
>> Evidently we can pilfer aosd06tmimpl, but re-reading what we
>> had there, it probably needs a lot of rewriting, so I have
>> not copied over the text.
>>
>> Regarding the authors list, I have made a guess at who is
>> involved this time. Evidently Julian and Pavel have supplied
>> 99.9% of the original ideas and all the hard work, so they
>> are the first authors; the rest of us follow in alphabetical
>> order. Of course others are welcome to join the effort.
>>
>> The #1 priority is to fill in the numbers in experiments.tex.
>> I've put in some descriptions already, and a table with a lot
>> of blanks.
>>
>> There are very rough drafts of intro etc in place, but don't
>> take these seriously yet. Without numbers there's little we
>> can say! We can however think carefully about what
>> experiments need doing to make it a coherent story, so that's
>> what I was aiming for with drafting this.
>>
>> I'll make some further changes to the outline over the
>> weekend, but then it would be good to have volunteers to take
>> charge of each section.
>>
>> We have ***only two weeks left*** until the paper deadline,
>> so this is *urgent*.
>>
>> -Oege
>>
>>
>>
>>
>>
>>
>
>
Received on Tue Mar 07 09:51:58 2006