On Fri, 12 Oct 2001, Sam TH wrote:
> I've spent a good bit of time this morning with Bugzilla, and I have
> some potentially interesting facts to report.
>
> Currently, we have 8 bugs marked 'dataloss'. These are the worst kind
> of bugs, they actually destroy people's work. They are:
>
> 665 1399 1437 1445 1467 1533 1689 2090
>
Could someone please verify that all these bugs still exist?
> 1 of these is still in the SUBMITTED state.
>
> We have 44 bugs marked 'crash'. They are:
>
> 484 654 773 774 804 867 915 970 974 977 1066 1101 1104 1161 1243 1253
> 1319 1397 1537 1685 1709 1726 1768 1778 1793 1808 1819 1859 1869 1885
> 1893 1932 1958 1964 1965 1971 2005 2009 2015 2017 2037 2069 2073 2100
>
I know many of these are invalid and relate to old binary builds. Could
some please look through these and remove the obviously incorrect ones?
> 15 of these are still in the SUBMITTED state.
>
> We have 25 bugs marked 'assert. They are:
>
> 595 773 1104 1111 1118 1123 1133 1161 1298 1364 1406 1445 1503 1538
> 1652 1657 1686 1726 1741 1814 1843 1919 1958 1965 2088
>
> 3 of these are still in the submitted state. Some of these overlap
> with those marked crash.
>
> These lists should be complete, at the moment, since I just went
> through and added keywords where neccessary.
>
> This is all out of 157 SUBMITTED bugs and 363 OPEN bugs.
>
> It seems clear to me that all the bugs marked 'dataloss' or 'crash'
> have to be fixed for 1.0. That's about 10% of the bug count
> currently. Therefore, to have a releaseable AbiWord by New Years
> means decreasing that count about 1 per day. This is not including
> other bugs we would like to fix.
>
> Personally, I think this argues for a feature freeze sooner rather
> than later. But draw your own conclusions.
>
Thanks Sam. I received an email from Alan saying we would prefer an
earlier feature freeze. Does anyone else agree?
Cheers
Martin
