Pricing ACL
in [Lisp]
|
Prev: Newlisp?
Next: cosx - sinx + 3x2
From: John Thingstad on 14 Dec 2005 06:30 On Wed, 14 Dec 2005 11:55:59 +0100, Pisin Bootvong <joesb.coe9(a)gmail.com> wrote: > John Thingstad wrote: >> On Wed, 14 Dec 2005 05:14:46 +0100, Pisin Bootvong >> <joesb.coe9(a)gmail.com> >> wrote: >> >> > I thought that if I bought MSVC++ and sell a product call >> > "start-up-vc.bat" which does nothing but start MSVC. And if I sells it >> > for 1$ and distributed MSVC++ with them, I can still get sued from MS. >> > And yet MS didn't charge MSVC per distributed app. >> > >> >> This is a bit off topic but MS-VC++ compiler tools are avaliable for >> free from microsoft and have been for a time now. >> It is the IDE they charge for. >> >> -- >> Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ > > > Most development environment out there don't sells the runtime. > They sell the development environment. They sells the convenient of > being able to edit/compile/debug. But they rarely sells the runtime. Not just libraries but also compiler and make and linker tools. > Franz's view is that if I deliver the Lisp image: > - My app will be capable of using "Compiler code, Eval". So what? Any > turing complete language is capable of writing an eval/interpreter > anyway. There is nothing special about eval, just a built-in > interpreter. Eval is just a convenient feature. Another of the > arguement is that "If we don't charge for each package someone will > build aproduct which do nothing but invoke the REPL and have all > feature of ACL and compete with us for less price". It is nonsense. > Can't you put in any License agreement to pervent that? Have you ever > seen that happens to LispWorks? It's the compiler thery are worried about not the evauator.. In particular if you have a compiler and save facillity you could deliver applications with their app with your code. It is not a ordinary senenario though. > - My app will have all the IDE features. Can't you disable this? You > can build a trial version that can't dump executable image and have > memory limit. But you can't disable IDE class to be built to exe? What > application mainly relied on ACL IDE component? > Yes you can disable IDE and GUI (I think that's what you mean) features. > Instead of requiring fee on special case of the product that use IDE > component, Franz suspect everyone will use it and charge on that > component. > > I think Franz target the server class environment anyway, so they don't > care about application that is sold in masses. > I'm not saying how this pricing model is good or bad for Franz, it's > their product after all. They can choose any pricing model they want. > All I'm saying is that I don't think this model will benefit Lisp in > any way. > I think Franz has set a standard for other Lisps. In this it has benefitted Lisp. For me comprehensive support is the primary factor why I chose Franz. I can't have my developers twiddle their thumbs for a week trying to figure out what to do. The most important cost to me is paying developers monthly. It takes a lot to make up for the cost of lost work hours. Software price is at best secondary. -- Using Opera's revolutionary e-mail client: http://www.opera.com/mail/
From: Kenny Tilton on 14 Dec 2005 10:15 Pisin Bootvong wrote: > John Thingstad wrote: > >>On Wed, 14 Dec 2005 05:14:46 +0100, Pisin Bootvong <joesb.coe9(a)gmail.com> >>wrote: >> >> >>>I thought that if I bought MSVC++ and sell a product call >>>"start-up-vc.bat" which does nothing but start MSVC. And if I sells it >>>for 1$ and distributed MSVC++ with them, I can still get sued from MS. >>>And yet MS didn't charge MSVC per distributed app. >>> >> >>This is a bit off topic but MS-VC++ compiler tools are avaliable for >>free from microsoft and have been for a time now. >>It is the IDE they charge for. >> >>-- >>Using Opera's revolutionary e-mail client: http://www.opera.com/mail/ > > > Yes i knew and I mean that whole IDE, not just the compiler that I > attached for 1$. Hmm. I have just downloaded VC++ 8 Express for free and registered and do not recall taking out my credit card. I admit I did not notice if there are indispensible add-ons I get only by buying some "Pro" version. OK, I also then downloaded the win32 sdk for free. Hmmm. Not that that is an add-on. > > Most development environment out there don't sells the runtime. > They sell the development environment. They sells the convenient of > being able to edit/compile/debug. But they rarely sells the runtime. > > Selling each runtime only reduced the chance of spreading the runtime, > therefore reduce the chance of application developed in such runtime of > reaching large user base. > However, Selling the runtime is still better than Franz's model of > price per an application distributed: If I sell my ACL-base-app to > someone who also have purchased another ACL-base-app from somewhere > else, it cost them twice for the "license to run ACL". > > Franz's view is that if I deliver the Lisp image: > - My app will be capable of using "Compiler code, Eval". So what? Any > turing complete language is capable of writing an eval/interpreter > anyway. ha ha ha, everyone loves to swing that "turing complete" club around. Would you like to write a C++ eval? Note that, like eval, it must cover the entire C++ language. There is nothing special about eval, just a built-in > interpreter. Eval is just a convenient feature. Convenient? No, for one Insanely Great application I did a while ago it was indispensible. In fact, my next product will rely on eval heavily as well. Indispensibly so. Oh, well, maybe I should soften up indispensibly -- an ugly and weaker alternative would be to load FASLs dynamically. Another of the > arguement is that "If we don't charge for each package someone will > build aproduct which do nothing but invoke the REPL and have all > feature of ACL and compete with us for less price". It is nonsense. > Can't you put in any License agreement to pervent that? Have you ever > seen that happens to LispWorks? > - My app will have all the IDE features. Can't you disable this? You > can build a trial version that can't dump executable image and have > memory limit. But you can't disable IDE class to be built to exe? What > application mainly relied on ACL IDE component? > > Instead of requiring fee on special case of the product that use IDE > component, Franz suspect everyone will use it and charge on that > component. > > I think Franz target the server class environment anyway, so they don't > care about application that is sold in masses. > I'm not saying how this pricing model is good or bad for Franz, it's > their product after all. They can choose any pricing model they want. > All I'm saying is that I don't think this model will benefit Lisp in > any way. > By "benefit" you must mean "help it grow by giving away stuff they spent millions to develop". No, it won't do that. Of course, that would not work since -- coming full circle -- they would not be in business very long. You all have a nasty "existence proof" to overcome: Franz is making money with Lisp. I think you all are missing something. Pricing when "early adopters" are your only market is always at a premium. You will not get hordes of buyers now matter how low the price, while early adopters will pay anything to get your product. Well, a premium anyway. Based on the blog traffic, we now see that Java has joined C++ on The Road to Obsolescence, and no statically-typed pretenders are to be had. The final round includes Python, Ruby, and -- holy prefix notation! -- Lisp. Let's bash Franz again when Apress acquires O'Reilly. Of course, who knows what their pricing will be then. kt
From: jayessay on 14 Dec 2005 10:58 alex.gman(a)gmail.com writes: > I wrote this as an email reply to R. Fateman, but I now feel like > posting it. (BTW, Richard, you probably meant to post your message in > the first place): So, now you've added "mind reader" to your claims of superior abilities. So, with this ability in your pocket and your professed amazing business acumen, why haven't you left MS and Bill Gates in the dust? Why isn't alex.inc (or whatever) in the fortune 50? /Jon -- 'j' - a n t h o n y at romeo/charley/november com
From: jayessay on 14 Dec 2005 11:25 Kenny Tilton <NOktiltonSPAM(a)nyc.rr.com> writes: > Pisin Bootvong wrote: > > - My app will be capable of using "Compiler code, Eval". So what? Any > > turing complete language is capable of writing an eval/interpreter > > anyway. > > ha ha ha, everyone loves to swing that "turing complete" club > around. Would you like to write a C++ eval? Note that, like eval, it > must cover the entire C++ language. Really. I sometimes wonder if people who say this sort of nonsense have any idea how foolish it makes them look. > > I'm not saying how this pricing model is good or bad for Franz, it's > > their product after all. They can choose any pricing model they want. > > All I'm saying is that I don't think this model will benefit Lisp in > > any way. I tell you one thing that greatly benefits Lisp that Franz has managed to accomplish: remain a viable business for 18 years. This is an important positive influenence when selling applications or proposals into major companies. > By "benefit" you must mean "help it grow by giving away stuff they > spent millions to develop". No, it won't do that. Of course, that > would not work since -- coming full circle -- they would not be in > business very long. > > You all have a nasty "existence proof" to overcome: Franz is making > money with Lisp. That's the primary problem with ideologues - they never let facts or reality get in the way of their beliefs. You watch - these people will now come up with a new set of "arguments" to "show" that Franz _can't_ actually exist. /Jon -- 'j' - a n t h o n y at romeo/charley/november com
From: Richard Fateman on 14 Dec 2005 11:53
Pisin Bootvong wrote: .... > > > Why would that make me more confident to sign the agreement that my > product, that contains Franz, instead of SBCL, doesn't violate anyone's > copyright? There are 2 parts to your product. The part you wrote yourself. If you deliberately copied it from something you didn't own, you could be in trouble. Then there is the part that Franz wrote. They protect you from issues about that part. That is part of what you pay them for. I don't know about SBCL. But any code covered by BSD is not warranted. When Berkeley was giving away 3BSD (UNIX for VAX, 68000), and the original Franz Lisp (different code base from Allegro), we were asked to sign all kinds of things by companies like IBM. Like UC Berkeley should assume all liability in case some of the 3BSD code was copyrighted by someone else. We didn't. For years IBM would not take 3BSD code. For some reason DEC was more willing to take our code. Maybe because their lawyers were not as careful or maybe they assumed that they would not be the target. IBM would be the target. Some government agencies required that none of 3BSD was produced from "convict labor". I wouldn't sign any agreement that some product I didn't > write myself doesn't violate any copyright, be it from SBCL, ACL or > Microsoft, would you? You might have trouble selling it. > And how likely do you think that Linux will > collapse in the next 5-10 year of copyright violation? Unfortunately, lawyers cost money whether the legal issues are real or not. So far as I can tell, IBM has plenty of lawyers and as long as they fight Caldera/SCO, the rest of us can just watch. If you are not on the same side of the table as someone with deep pockets, you might get sued. > ...... >> >>Well, when I ask for help from a vendor of Windows-based software, I >>am often told that I should mess around with the operating system, get >>a download of something, etc. The vendors rely on Microsoft maintaining >>a certain presence. We may argue that MS doesn't do everything we want >>it to do, but it usually doesn't require downloading fresh source code >>for the whole operating system, editing a makefile, and recompiling everything. >>That's the kind of advice I get if cygwin is broken. >> > > > Right, but you still get such advice and support from the product X > first. Correct. You don't go directly to MS support and ask why Some X software > doesn't work, do you? Your Tahiti situation was that the product X > vendor is not willing to provide any support and direction. Even if > that irresponsible vendor use ACL, I still wouldn't expect any better > support. Incorrect. Say the vendor determines that the bug is in the Lisp. Who fixes it? The vendor? The open-source community? The vendor may not be (should not have to be ...) an expert in Lisp implementation. Could he fix a bug in (say) the garbage collector for GCL? In the case of a commercial Lisp, there is at least a prospect of getting professional help. Nevertheless, I think that if you find a bug in Franz software you are likely to get a fix or a workaround. From the open source community you also might get a fix or a workaround, but maybe not. It depends on how "shallow" the bug is, how easy to reproduce, who is looking for things to do, how many people take pride in their volunteerism, how many people total there are in the community and how many are really expert as opposed to just hacking. There are lots of people in the linux community. Many fewer in GCL. > ...... > >>There are ways of auditing the sale of products. While MS might not >>catch you making an extra copy of Word on your laptop, you can be sure >>that if a large organization, even a university, buys software, >>(or sells software) it cannot use piracy for long. (Unless it is in >>a country that does not respect patents, copyrights, etc.) >> > > > Right, that's why you don't have to be thinking of "Product Y that > provide all of ACL and sell for less" kind of situation if you put the > restriction in your EULA. I don't understand this. What are you saying? 1.As long as your product costs more than ACL it is ok to include all of ACL? (and also pay no royalties to Franz?) Because no one would buy it instead of ACL? [problem: what if you don't enforce your EULA, there are all these ACLs distributed, hurting you and Franz.] 2. If your product costs less than ACL it is OK to include all of ACL because.. um, someone would be giving you money and um, that is good? 3. You are proposing that as long as the product does not contain every single bit of ACL, it is ok to redistribute without royalties to Franz? [Say, removing Roman Numerals from the formatted output...] or that there should be some negotiation with Franz about royalties. {I gather this is the current pricing policy}. Auditing, which you seem to favor, means that the EULA from Franz and also from you will contain a royalty provision which also allows Franz to audit (a) your books; (b) your customers' books (c) if they also produce software, their customers' books. etc. >..... >> > > > May be something like VMWare player, say AllegroPlayer. > Just the runtime, where anyone can install for free and use for any > purpose. > Just the pure runtime, no IDE, no interface at all, without ability to > compile file, and must run from fasl. No AllegroCache or any of that > Allegro addon that you think deserve to be charge per copies. One problem with Lisp is that the runtime for Lisp tends to be almost all of the Lisp. Symbolics had a delivery system on IBM-PCs called CLOE, (common lisp object environment) that it used for "products". The only one I ever saw was Macsyma, but maybe there were others. CLOE/Macsyma had a compiler since Macsyma uses compilation. Macsyma also has an "execute lisp command" capability. So the delivery system could leave out the Symbolics IDE, but had to leave in, ANSI Common Lisp. I guess it is a business and technical decision for Franz as to whether they should design and develop a product like "AllegroPlayer" and then give it away. Such a product would not be free to develop, support, document. Would this increase sales of development systems enough to justify the cost? (Franz already gives out (source) for a variety of programs, and also gives out a free trial Allegro version.) RJF > |