Re: [racket-dev] Racket 5.3 pre-release impressions?
Thanks, Doug. From talking with a few people, it sounds like 5.3 is shaping up pretty normally for a release, and the releases have been high-quality. I was just a little spooked by running into two bugs very quickly (two points determine a line, after all), but I haven't found any since those. Neil V. _ Racket Developers list: http://lists.racket-lang.org/dev
[racket-dev] Racket 5.3 pre-release impressions?
Regarding Racket 5.3, I am more cautious than I recall being about a previous Racket minor version release. The information I have so far is mixed, rather that overwhelmingly reassuring. If anyone has comments on their sense of 5.3 reliability at this point, that might help me. Some good news: one of my consulting clients had me do some preliminary testing of their large code base with the Racket 5.3 pre-release, and there has been no problem with that testing so far. However, in testing 5.3 pre-releases with my personally-owned code, rather than clients' code, so far I've found what looks like 2 bugs in the 5.3 pre-release. This was surprising to me, since I expected 0 bugs. I have not yet tried the majority of my personally-owned code. Until I get a better sense, I would expect to keep production servers at 5.2.1 for a while after the 5.3 release. I'm also probably going to go to some trouble to avoid using new 5.3 features in some code that might be deployed to production in the very near term. More information might be reassuring to me. Neil V. _ Racket Developers list: http://lists.racket-lang.org/dev
Re: [racket-dev] Racket 5.3 pre-release impressions?
FWIW, I just tested 16 or so additional PLaneT packages in DrRacket 5.3 pre-release, and no problems. Neil Van Dyke wrote at 07/28/2012 02:56 PM: Regarding Racket 5.3, I am more cautious than I recall being about a previous Racket minor version release. The information I have so far is mixed, rather that overwhelmingly reassuring. If anyone has comments on their sense of 5.3 reliability at this point, that might help me. Some good news: one of my consulting clients had me do some preliminary testing of their large code base with the Racket 5.3 pre-release, and there has been no problem with that testing so far. However, in testing 5.3 pre-releases with my personally-owned code, rather than clients' code, so far I've found what looks like 2 bugs in the 5.3 pre-release. This was surprising to me, since I expected 0 bugs. I have not yet tried the majority of my personally-owned code. Until I get a better sense, I would expect to keep production servers at 5.2.1 for a while after the 5.3 release. I'm also probably going to go to some trouble to avoid using new 5.3 features in some code that might be deployed to production in the very near term. More information might be reassuring to me. Neil V. _ Racket Developers list: http://lists.racket-lang.org/dev
Re: [racket-dev] Racket 5.3 pre-release impressions?
I've tried all of my packages and had one major snag that Matthew thinks he has fixed. For my large simulations, it seems to execute faster, but that is just an impression. Many of them are actually constrained by quick and dirty plots that may be rendering tens or hundreds of thousands of points. Doug On Sat, Jul 28, 2012 at 12:56 PM, Neil Van Dyke n...@neilvandyke.org wrote: Regarding Racket 5.3, I am more cautious than I recall being about a previous Racket minor version release. The information I have so far is mixed, rather that overwhelmingly reassuring. If anyone has comments on their sense of 5.3 reliability at this point, that might help me. Some good news: one of my consulting clients had me do some preliminary testing of their large code base with the Racket 5.3 pre-release, and there has been no problem with that testing so far. However, in testing 5.3 pre-releases with my personally-owned code, rather than clients' code, so far I've found what looks like 2 bugs in the 5.3 pre-release. This was surprising to me, since I expected 0 bugs. I have not yet tried the majority of my personally-owned code. Until I get a better sense, I would expect to keep production servers at 5.2.1 for a while after the 5.3 release. I'm also probably going to go to some trouble to avoid using new 5.3 features in some code that might be deployed to production in the very near term. More information might be reassuring to me. Neil V. _ Racket Developers list: http://lists.racket-lang.org/dev _ Racket Developers list: http://lists.racket-lang.org/dev
