- "Programming language life expectancy" - 2 Updates
- [OT] free will - 2 Updates
- On the road again - 3 Updates
- getting system info in c++ linux - 1 Update
- Faster than Boost, Cereal and Protobuf - 1 Update
Lynn McGuire <lynnmcguire5@gmail.com>: Aug 30 08:08PM -0500 "Programming language life expectancy" https://www.johndcook.com/blog/2017/08/19/programming-language-life-expectancy/ "The Lindy effect says that what's been around the longest is likely to remain around the longest. It applies to creative artifacts, not living things. A puppy is likely to live longer than an elderly dog, but a book that has been in press for a century is likely to be in press for another century." ""We don't know what language engineers will be coding in in the year 2100. However, we do know that it will be called FORTRAN." — C.A.R. Hoare" Lynn |
"Mr. Man-wai Chang" <toylet.toylet@gmail.com>: Aug 31 09:54PM +0800 On 31/8/2017 9:08 AM, Lynn McGuire wrote: > ""We don't know what language engineers will be coding in in the year > 2100. However, we do know that it will be called FORTRAN." — C.A.R. Hoare" > Lynn No mention of xBase and its dialects... Sad. No COBOL as well! -- @~@ Remain silent! Drink, Blink, Stretch! Live long and prosper!! / v \ Simplicity is Beauty! /( _ )\ May the Force and farces be with you! ^ ^ (x86_64 Ubuntu 9.10) Linux 2.6.39.3 不借貸! 不詐騙! 不援交! 不打交! 不打劫! 不自殺! 請考慮綜援 (CSSA): http://www.swd.gov.hk/tc/index/site_pubsvc/page_socsecu/sub_addressesa |
Rod Pemberton <NeedNotReplyHere@xrsevnneqk.cem>: Aug 31 12:09AM -0400 On Wed, 30 Aug 2017 16:46:48 +0100 > Yes you are a fucking child mate. Says the guy hiding behind the name of a Red Dwarf puppet. Do you live in your parents basement too? Mate? I'm definitely not your sex partner. Apparently, British have a different meaning for "mate" than Americans. As the meaning of your words are bizarre to me, I'm really somewhat confused when it comes to British swearing ... Does insulting me make you an arsehole, wanker, bellend, dickhead, or a twat? It's really hard to decide from the descriptions. You seem a bit effeminate, so I'm partial to that last one. It plays well in American English too. But, as a munter Billy no-mates muppet chav, it's clearly your choice as to which one you're called. Rod Pemberton -- Isn't anti-hate just hate by another name? Isn't anti-protesting just protesting by another name? Peace is a choice that both sides rejected. |
leigh.v.johnston@googlemail.com: Aug 31 12:08AM -0700 Take your medication mate. /Flibble |
David Brown <david.brown@hesbynett.no>: Aug 30 08:33PM +0200 >> God bothering in any form is unwanted here. > Leigh, a kind word here would be to say congratulations on this > milestone and best wishes in the years ahead. If everybody made a self-congratulatory post for every birthday of the company where they work, the newsgroup would be as useless as Facebook. If you make a big new release, or suddenly get serious interest for your software from someone big, then maybe it would be interesting. But no one here cares if your company is 18 years old or 19 years old - it is utterly irrelevant to anyone but you. It is not a milestone of any significance to the outside world. And yes, your post is an excuse for religious drivel and nothing to do with C++. What you believe is up to you - religious freedom is a key human right. But you can talk about that in religious newsgroups, not a C++ group. |
woodbrian77@gmail.com: Aug 30 12:20PM -0700 On Wednesday, August 30, 2017 at 1:33:52 PM UTC-5, David Brown wrote: > one here cares if your company is 18 years old or 19 years old - it is > utterly irrelevant to anyone but you. It is not a milestone of any > significance to the outside world. Sorry if you don't appreciate my rehashing things, but people need tools that will outlive their projects. The idea of private property (closed source) is solid and will help the company for years to come. > And yes, your post is an excuse for religious drivel and nothing to do > with C++. I'm asking for ideas on how to improve my software and documentation. Brian Ebenezer Enterprises - Enjoying programming again. http://webEbenezer.net |
David Brown <david.brown@hesbynett.no>: Aug 31 12:11AM +0200 > need tools that will outlive their projects. The idea of private > property (closed source) is solid and will help the company for > years to come. I can't make sense of that paragraph - sorry. >> with C++. > I'm asking for ideas on how to improve my software and > documentation. Ask away - but do so without the preaching. When you talk about C++, or ask about C++, that's fine. Just don't bring religion into it. (If you feel some god guides your programming, that's no problem to me - but it is a matter between you and him, not a matter for a technical newsgroup.) |
"Chris M. Thomasson" <invalid@invalid.invalid>: Aug 30 01:55PM -0700 On 8/27/2017 1:52 PM, Chris M. Thomasson wrote: > It is very basic information, and not going to help wrt creating a > topographic map of NUMA nodes. We basically need some OS specific > information here. Just brainstorming here, fwiw: Since C++ has std::thread::hardware_concurrency, well, what about thinking of creating something like: std::thread::hardware_concurrency_tree That returns information in the form of a tree structure that describes, or encodes the way the system actually arranges the number of cpu's returned from std::thread::hardware_concurrency? You basically get a root node that has all of the NUMA nodes in the "system". So, crudely and quickly, for instance, if there are 2 NUMA nodes, it gives you a linked list NL with two items, each of which are a NUMA node. Each node in NL represents a CPU item; each CPU item has multiple nodes representing its virtual cores VCPU; each VCPU has its "hyper-threading" cores on each VCPU; and down to per-thread... Afaict, this is zooming down from the high level at 2 NUMA nodes, into the per-thread leaves. The "fractal" zoom ends in a tree where terminating leaves are thread nodes. I guess we can go one level deeper and add n fibers to each per-thread leaf, transform the thread leaf into a fiber leaf, in a sense. Humm... The overall layout for NUMA down to thread and/or fiber has always sounded sort of "fractal in nature" to me. My crude rational for a possible proposal would be: We have std::thread::hardware_concurrency. This is good information, however its not very useful wrt mapping out the system. An additional and optional function just might be in order: std::thread::hardware_concurrency_tree, or something. That gives deeper information on how the result of std::thread::hardware_concurrency is arranged. Also, what about: std::thread::hardware_memory_tree that gives various cache line sizes and how they are padded, and arranged in the NUMA tree. Combine this with optional thread affinity masking, and a program in C and/or C++ can combine this with standard atomics to create highly scaleable, low level, systems. Imvvvho, something like: std::thread::hardware_concurrency_tree is probably "low-level enough" for C or C++ wrt providing aid to a programmer that wants to use threading and atomics effectively, and continue to use the standard wrt gaining the cache line size to properly pad and align their structures in memory: We do not want any false-sharing, damn it! ;^) |
Dombo <dombo@disposable.invalid>: Aug 30 08:19PM +0200 Op 21-Aug-17 om 6:43 schreef Daniel: > https://github.com/danielaparker/jsoncons Slightly off-topic; a year ago I was looking for a C++ JSON library for a quick proof-of-concept. Though there are plenty to choose from I didn't find any I liked. So I wrote my own and considered putting it on GitHub if I ever got around to polish it a bit. The link above made me curious what you came up with. For some reason this library hasn't popped up on my radar before. I must say I really liked what I saw. In many ways your library is very similar to what I created, it also has a couple of things that I planned to do if I had the time (or need) and things I hadn't even considered. If I had found your library a year ago I most likely wouldn't have bothered to create my own, but if I ever need a JSON library again I know where to look now. Nice work! |
You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page. To unsubscribe from this group and stop receiving emails from it send an email to comp.lang.c+++unsubscribe@googlegroups.com. |