- Division by zero - 3 Updates
- Complex std::map Key - Possible? - 1 Update
- Another difference between gcc and clang - 2 Updates
| David Brown <david.brown@hesbynett.no>: Nov 01 09:56PM +0100 On 01/11/16 16:38, Jerry Stuckle wrote: >> - before Wikipedia was invented.) > I never said you claimed to be an expert on these sorts of mathematical > structures. I didn't say you said anything like that - that's why I put in brackets "for the record", in case /anyone/ had thought I was an expert on these points. We have had enough arguments and disagreements in the past - this was not intended to be one - I agree with you here! > And there are facts, and there is Wikipedia. Sometimes the two agree. > But I do agree, advanced math articles are typically written by people > who know what they are talking about. Unlike many Wikipedia articles. Yes, but in this case the articles are fine (though they are not designed to be maths tutorials for people new to these topics). I agree that sometimes Wikipedia articles can be wrong - and sometimes surprisingly so even on simple factual matters. It all depends on who wrote the article, how careful and serious they are, whether they have some extra agendas or biases, and how many others read, edit and correct the article. Usually the more technical articles are quite accurate, while things like political topics are more likely to start out as opinion pieces. > I've even seen Dubai claimed as the capital of the UAE (it's Abu Dhabi). > How could someone get THAT one wrong? That sounds like a genuine mistake - and not an uncommon one. If it was on an article about the UAE, politics or geography, then it should definitely not have occurred. It is a little more excusable if it were merely mentioned in passing in another article. But there is no doubt that mistakes like this happen in Wikipedia, and one must be aware of the possibility when using it as a reference or guide. It is also interesting sometimes to look at the same article in different languages. I looked up "The Battle of Largs" once. This was a fairly sizeable battle in which the Scottish king finally beat the Vikings and forced them out of Scotland (or at least, forced them out of power in Scotland). The English language version was quite extensive, and described it as a large and important battle with serious political effects on the history of Scotland. The Norwegian version, on the other hand, described it as a minor scuffle of no importance. The contrast between the viewpoints of the winners and the losers was clear! |
| Paavo Helde <myfirstname@osa.pri.ee>: Nov 01 11:57PM +0200 On 1.11.2016 21:40, Mr Flibble wrote: > You obviously have a different idea as to what mathematics actually is > compared to the rest of us. Dividing by zero is undefined, period. Count me out of the "rest of us", please. Studied theoretical physics too long. Thanks! Paavo |
| "Rick C. Hodgin" <rick.c.hodgin@gmail.com>: Nov 01 03:16PM -0700 On Tuesday, November 1, 2016 at 5:13:08 PM UTC-4, David Brown wrote: > omitting hardware division is a reasonable tradeoff for the smallest and > cheapest devices. In terms of the number of devices produced, these are > the sizeable majority. I disagree with that goal. The software written for these devices is typically not very complex or large (due to the constraints of the devices). On the other hand, Android-like devices and larger have many Gigabytes of memory and storage, and run much larger much more powerful applications. It should be expected for any language designer to target a growing set of internal abilities exposed to it through the ISA, not less, and not a decreasing set. Hardware is very cheap these days. The fact that it's still cheaper in the smaller form factors is of very little overall consequence because they have significantly lesser demands of a language, even enough that many of them could be handled with simple assemblers with supporting function call libraries provided for by the manufacturers just as easily as relying upon something as complex as a C standard library. In fact, I'd wager that in such applications the C standard library would be a huge hindrance due to the special needs of limited hardware for optimization. > IIRC, hardware division was dropped from some members of the 68k family > because someone noticed that a software division loop was faster than > the hardware instruction! I see the future as an increasing set of abilities exposed to the software developer, up to a point, but always including the base and fundamental operations in integer and fp. Best regards, Rick C. Hodgin |
| Paavo Helde <myfirstname@osa.pri.ee>: Nov 01 11:51PM +0200 On 1.11.2016 23:06, Vir Campestris wrote: > Welcome to template error messages. They can be > _horrible_ - hundreds of characters long. You meant hundreds of lines. |
| Juha Nieminen <nospam@thanks.invalid>: Nov 01 07:06AM This piece of code, by itself, does not cause a compiler error with gcc6, but does with clang: //------------------------------------------- template<typename T> class C { public: test(int, int); }; //------------------------------------------- clang says: error: C++ requires a type specifier for all declarations gcc says nothing and just accepts it. I wonder what it's thinking. --- news://freenews.netfront.net/ - complaints: news@netfront.net --- |
| "Alf P. Steinbach" <alf.p.steinbach+usenet@gmail.com>: Nov 01 11:47AM +0100 On 01.11.2016 08:06, Juha Nieminen wrote: > clang says: > error: C++ requires a type specifier for all declarations > gcc says nothing and just accepts it. I wonder what it's thinking. It's thinking "implicit int". That's never been part of C++. Cheers!, - Alf |
| 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. |
No comments:
Post a Comment