- Boost-users - lists.boost.org

Re: [Boost-users] Ownership of Boost IP?
by Nat Goodspeed 04 Aug '06

04 Aug '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of Paul Giaccone > Sent: Friday, August 04, 2006 7:38 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] Ownership of Boost IP? > > Nat Goodspeed wrote: > > I suppose the next round would involve my arguing that it would take me > > many months to implement even part of the functionality in question > > > The answer to that would be to contact the Boost developers for all the > libraries you intend to or might use, and ask them how long it took them > to code up their respective libraries. [Nat] I'm hoping to avoid any approach that would add per-Boost-library overhead. Having to decide in advance the set of libraries I intend to use - and later facing additional internal battles to get permission to use one more -- would be discouraging. But I take your point: you're talking about an example set, not literally suggesting that I repeat the process every time I desire to incorporate another new Boost library. > Add it all up and present that > figure to your boss, pointing out that these people were at least as > competent as you or any other programmer he might care to employ. Does > he *really* want to wait a year or two for the code you are writing for > him? [Nat] Certainly he wouldn't want that. I guess I wasn't thinking clearly last night. The real problem is that nobody -- not me, not my boss, not the lawyers -- wants me to (attempt to) reimplement any Boost library. If we can't use Boost, then we must resort to the ugly hand-coded mechanisms that C++ programmers used before Boost emerged. No smart pointers? Delete the !@#? pointers yourself! No bind()? No boost::function? Quit griping and use the STL adapters! Can I make a case that the small incremental costs in coding time and reliability would add up to a bigger hit than the hypothetical risk of lawsuit? It seems doubtful, since all the factors involved in such an argument are sheer guesswork -- and every party will believe their own guesses rather than their opponents'. My boss might even be sympathetic. But my boss isn't the real problem here. > You might also want to ask your boss whether he is concerned that > Microsoft (or whoever wrote your compiler) might sue for use of their > version of C++, or whether he thinks Bjarne Stroustrup might be sending > you a writ... Of course, that would just be ridiculous, wouldn't it? ;) [Nat] Microsoft it is. I think I heard that Microsoft's licenses (purport to) protect you, the user of their products, from this kind of liability. That may be what raised this whole issue in our legal dept. in the first place.

2 1

Re: [Boost-users] Ownership of Boost IP?
by james.jones＠firstinvestors.com 04 Aug '06

04 Aug '06

From: "Nat Goodspeed" <ngoodspeed(a)solidworks.com> > As the lawyers' position is "Don't use > Boost," I can't see them making that effort, so it would be up to me (or > other interested developers) to pursue it. Which leaves me/us vulnerable > to charges of "wasting time" that could "better" be spent reinventing > Boost technologies. How about mentioning that a bunch of Boost stuff (ideas, at least) is intended for use in TR1 and, AFAIK, TR2 and TRn. And that surely standard libraries wouldn't be subject to these IP concerns? (Of course, they might reply: "Fine, once it gets into standard libraries you can use it." Which means you're waiting years.) - James Jones Administrative Data Mgmt. Webmaster 375 Raritan Center Pkwy, Suite A Data Architect Edison, NJ 08837

4 3

Re: [Boost-users] Ownership of Boost IP?
by Sohail Somani 04 Aug '06

04 Aug '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of Robert Ramey > Sent: Friday, August 04, 2006 9:52 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] Ownership of Boost IP? > > > Realistically, this situation isn't about the facts - its about > someone having authority beyond his expertise. Its a > political issue not a technical one. Boost and its developers > can only do so much. Therefore, it can all be solved with a nice lunch with the boss! Bosses are good at this stuff. Atleast the good ones :)

1 0

Re: [Boost-users] Ownership of Boost IP?
by Nat Goodspeed 04 Aug '06

04 Aug '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of Robert Ramey > Sent: Friday, August 04, 2006 11:18 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] Ownership of Boost IP? > > Nat Goodspeed wrote: > > My employer's legal department has just stated that we may not use > > Boost for product development because of a hypothetical lawsuit risk. > > Hmmm - wouldn't his same concern apply to any software you > don't write yourself? Suppose you use VC 7.1. It comes with > dinkumware libraries - does your legal dept require a release > from dinkumware?. Assuming it does, how does assure itself > that some dinkumware developer didn't use some code/knowledge > from his previious job or was inspired by some snippet he found > on the net. OK so you get releases and hold harmless clauses > from all the past dinkumware developers. Now how about the > windows API - after all that's just another library. Are you going > to use calls from that? Does your legal department feel confident > that no microsoft employee or manager has incorporated code > from other sources? If it does, its pretty clear they've not aware > of all the lawsuits microsoft has had in this regard. And the fact > is that none of the lawsuits have resulted in damages to microsoft > customers or users or developers' who have leveraged on such > code. [Nat] To play devil's advocate for a moment -- I'm sure the legal dept. would say that's because Microsoft promises protection from such actions, in return for the license fee. If I could show precedent where someone attempted to sue users of (or developers who integrated) an Open Source project -- but lost -- that would be a stronger argument. > Such a decision has technical and economic implications far, far > beyond what your legal department (and probably management - if its gotten > this far) are aware of. [Nat] No argument from me there. > Maybe more practical advice would be > to suggest to your boss that he get a second legal opinion. [Nat] Interesting suggestion! I'll see if I can make anything happen along those lines.

2 1

Re: [Boost-users] shared_ptr in API
by Sohail Somani 04 Aug '06

04 Aug '06

If the pointer is meant to be owned by the caller, you could consider references (assuming the objects are required of course). > -----Original Message----- > From: boost-users-bounces(a)lists.boost.org > [mailto:boost-users-bounces@lists.boost.org] On Behalf Of > Mikhail Kubzin > Sent: Thursday, August 03, 2006 3:45 AM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] shared_ptr in API > > Did you consider using weak_ptr in your API? > > > > _______________________________________________ > Boost-users mailing list > Boost-users(a)lists.boost.org > http://lists.boost.org/mailman/listinfo.cgi/boost-users >

1 0

permutation_iterator: No conversion support
by Daniel Krügler 04 Aug '06

04 Aug '06

Hello, I observed this weekend, that permutation_iterator cannot provide compatible conversions, in constrast to all other iterators. The problem is due to access right violations in the "conversion" pseudo copy c'tor: template<class OtherElementIterator, class OtherIndexIterator> permutation_iterator( permutation_iterator<OtherElementIterator, OtherIndexIterator> const& r , typename enable_if_convertible<OtherElementIterator, ElementIterator>::type* = 0 , typename enable_if_convertible<OtherIndexIterator, IndexIterator>::type* = 0 ) : super_t(r.base()), m_elt_iter(r.m_elt_iter) {} This c'tor cannot be instantiated successfully, even if SFINAE has allowed that: The private member m_elt_iter cannot be accessed from another permutation_iterator<OtherElementIterator, OtherIndexIterator> and the result is a compile error. One should either provide friendship to permutation_iterator<typename, typename> or the private member should be readable via a public member function to fix this. I have provided a corresponding bug report http://sourceforge.net/tracker/index.php?func=detail&aid=1530999&group_id=7… It would be very nice, if that fix could go into boost 1.34! Many thanks and Greetings from Bremen, Daniel Krügler

2 1

MS VC2005 + Boost Setup
by Heiko Günther 04 Aug '06

04 Aug '06

Hello! I have some trouble to setup the Boost library with VC++ 2005. I am rather new to the library, so I could need some help. I have downloaded Version 1.33 and extracted the files to location A ("d:\programs\boost" in my case). Visual Studio is located in location B ("d:\programs\VS"). What is the best way to build the library files? I especially need the compiled library of the "signal" class. Since I have no experience with this bjam tool any help would be appreciated. I can't figure out how this works using the information from the project's website.. Best regards, Heiko

4 4

Re: [Boost-users] Ownership of Boost IP?
by Nat Goodspeed 04 Aug '06

04 Aug '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of John Maddock > Sent: Friday, August 04, 2006 5:09 AM > To: boost-users > Subject: Re: [Boost-users] Ownership of Boost IP? > > Whew, of course now I feel like no one should write code, and I should go > look for another profession! But hopefully there is some ammunition in > there you can use. [Nat] Thank you!! I will certainly give it a try!

1 0

Re: [Boost-users] Ownership of Boost IP?
by Nat Goodspeed 04 Aug '06

04 Aug '06

> -----Original Message----- > From: boost-users-bounces(a)lists.boost.org [mailto:boost-users- > bounces(a)lists.boost.org] On Behalf Of Gennaro Prota > Sent: Thursday, August 03, 2006 9:13 PM > To: boost-users(a)lists.boost.org > Subject: Re: [Boost-users] Ownership of Boost IP? > > On Thu, 3 Aug 2006 19:39:17 -0400, "Nat Goodspeed" wrote: > > >My employer's legal department has just stated that we may not use Boost > >for product development because of a hypothetical lawsuit risk. > > Devin Smith, the lawyer who wrote the Boost Software License, > anticipated these issues; see "Benefit of Common Software License" in: > > <http://www.boost.org/more/license_info.html> > > where he also mentions a possible way to address the problem. [Nat] I think you mean this passage? "But this can be addressed, not by trying to craft the perfect standard license, but by informing the corporations that they can, if they don't like the standard license, approach the authors to negotiate a different, perhaps even paid, license." Thank you, I hadn't previously found the cited discussion. But... If that's the best I can put forth, then I worry that our legal dept. will assert that tracking down individual library authors and negotiating with each of them will largely negate the savings in software development time. As the lawyers' position is "Don't use Boost," I can't see them making that effort, so it would be up to me (or other interested developers) to pursue it. Which leaves me/us vulnerable to charges of "wasting time" that could "better" be spent reinventing Boost technologies. I suppose the next round would involve my arguing that it would take me many months to implement even part of the functionality in question -- and the legal staff asking whether my boss should consider someone more competent instead. :-/ Any additional ammunition would be gratefully received...

4 3

How do I use regex to match two consecutive newlines?
by Tommy Li 04 Aug '06

04 Aug '06

I've tried matching like this: std::string input = filecontents; boost::regex re_paragraphs("\\n\\n"); boost::sregex_token_iterator input_token_it(input.begin(), input.end(), re_paragraphs, -1); boost::sregex_token_iterator end_token_it; for( ; input_token_it != end_token_it; input_token_it++ ) { output += "TOKEN FOUND" } return output; But this never works. It matches when I use a part of a word as a token. The problem is that it's just not matching the newlines. I've looked into the input's hex and there are indeed two newline chars consecutively. I think I'm missing something obvious. Anyone have any idea?

3 6