Can extra processing threads help in this case?
in [MFC]
|
From: Hector Santos on 31 Mar 2010 16:28 Pete Delgado wrote: > "Peter Olcott" <NoSpam(a)OCR4Screen.com> wrote in message > news:bsGdnZJPcN687C7WnZ2dnUVZ_vednZ2d(a)giganews.com... >> If the only thing that I must insure does not occur is that the customer >> is not charged for a transaction that did not complete, this simplifies >> the design. For every complex set of issues there is always a way to >> minimize the effective complexity. >> >> One thing that I need to know hopefully without the need to read a 1,000 >> page book > > Had you taken the time to read portions of the 1,000 page books and various > articles that were suggested much earlier in this thread, you might have > saved yourself a lot of requirements thrashing and the embarassment of > looking ignorant when people suggested various solutions to you. > > I realize that not everyone can read hefty technical tomes and come away > with an understanding of the topic, but in many cases sitting down with a > good book on software development can pay dividends in years to come. Thus > far I haven't seen anything suggested to you that isn't well written, > interesting and full of good information. Why not give it a try??? Peter D. I think its is pretty clear the problem Peter 0. has. He is trying to write a patent for a guarantee 100 TPS, 100ms response process for his OCR system designed and integrated under certain contraints; Pure memory operations, short distances between end points, fault tolerance, etc. However, he is learning that his patent will have no merit if he can not guarantee it under these constraints. He can't believe or use any part of what people are telling him, including whatever is said in books, MSDN or any other technical article. He needs to provide an "abstract view" that it is at the very least possible for his patent to have any substance or validity. The implementation details are not necessary. So in other words, if anyone else creates a OCR web service and its faster than his, he will claim it infringes on his patent and trade secrets because the methods used can only be done using pure memory and short distances between end points. He can not believe that its possible for a process to naturally use the optimized OS VM technology and be written to be as fast or better than his own pure memory access process. Thats the problem here. Its easy to see through this simplistic patent troll. -- HLS
From: Pete Delgado on 31 Mar 2010 16:42 "Peter Olcott" <NoSpam(a)OCR4Screen.com> wrote in message news:i5ednevFTopcOC7WnZ2dnUVZ_rednZ2d(a)giganews.com... > I can only do that to the extent that it is cost-effective. I am doing > that now on those things that I must know in depth, so far that is three > books totaling 2500 pages. The reason that I come to the various > newsgroups is to reduce the search time. I could simply read a couple > dozen 1000 pages books and skip this newsgroup step. I think that you are missing the point that many of the references cited will not only help with your immediate problems but also those you are likely to face in the immediate future and those that you are currently unaware of. Just out of curiosity, what books are you reading at the moment? > > When I ask whether or not there is a way to verify with certainly that an > email message has been received, providing a yes or a no does not take > that much effort. referring me to the current working set of email > specifications is not very helpful. The spec would provide the answer that you were looking for as would a search of this thread if memory serves me correctly. Granted, neither method provides the answer in the quick, concise format that you desire but then again it seems to me that the vast majority of your design decisions were made using a "Magic 8 Ball" technique anyway and thus are not based upon anything but fantasy to begin with. Why not cut out the step of posting in this newsgroup and ask your question here? http://web.ics.purdue.edu/~ssanty/cgi-bin/eightball.cgi BTW: I asked the Magic Eight Ball if "Peter Olcott will ever finish his OCR product" and the answer given was "My sources say no." YMMV -Pete
From: Peter Olcott on 31 Mar 2010 17:32 "Pete Delgado" <Peter.Delgado(a)NoSpam.com> wrote in message news:ujfW9KR0KHA.260(a)TK2MSFTNGP05.phx.gbl... > > "Peter Olcott" <NoSpam(a)OCR4Screen.com> wrote in message > news:i5ednevFTopcOC7WnZ2dnUVZ_rednZ2d(a)giganews.com... >> I can only do that to the extent that it is >> cost-effective. I am doing that now on those things that >> I must know in depth, so far that is three books totaling >> 2500 pages. The reason that I come to the various >> newsgroups is to reduce the search time. I could simply >> read a couple dozen 1000 pages books and skip this >> newsgroup step. > > I think that you are missing the point that many of the > references cited will not only help with your immediate > problems but also those you are likely to face in the > immediate future and those that you are currently unaware > of. > > Just out of curiosity, what books are you reading at the > moment? > In order of priority HTTP The Definitive Guide Advanced Programming in the Unix Environment UNIX Network Programming The Sockets Networking API MySQL Developers Library >> >> When I ask whether or not there is a way to verify with >> certainly that an email message has been received, >> providing a yes or a no does not take that much effort. >> referring me to the current working set of email >> specifications is not very helpful. > > The spec would provide the answer that you were looking > for as would a search of this thread if memory serves me > correctly. Granted, neither method provides the answer in > the quick, concise format that you desire but then again > it seems to me that the vast majority of your design > decisions were made using a "Magic 8 Ball" technique > anyway and thus are not based upon anything but fantasy to > begin with. Why not cut out the step of posting in this > newsgroup and ask your question here? > http://web.ics.purdue.edu/~ssanty/cgi-bin/eightball.cgi > > BTW: I asked the Magic Eight Ball if "Peter Olcott will > ever finish his OCR product" and the answer given was "My > sources say no." > Time will tell. I have 20,000 hours invested so far. > YMMV > > -Pete >
From: Pete Delgado on 31 Mar 2010 18:50 "Peter Olcott" <NoSpam(a)OCR4Screen.com> wrote in message news:6Z6dnbu7zLjhIi7WnZ2dnUVZ_uOdnZ2d(a)giganews.com... >> The spec would provide the answer that you were looking for as would a >> search of this thread if memory serves me correctly. Granted, neither >> method provides the answer in the quick, concise format that you desire >> but then again it seems to me that the vast majority of your design >> decisions were made using a "Magic 8 Ball" technique anyway and thus are >> not based upon anything but fantasy to begin with. Why not cut out the >> step of posting in this newsgroup and ask your question here? >> http://web.ics.purdue.edu/~ssanty/cgi-bin/eightball.cgi >> >> BTW: I asked the Magic Eight Ball if "Peter Olcott will ever finish his >> OCR product" and the answer given was "My sources say no." >> > > Time will tell. I have 20,000 hours invested so far. Assuming that your 20,000 hours figure is correct, from my calculations, if you had worked 8 hours a day, 7 days a week, 365 days a year (not counting leap days which I will give you as a holiday for your dilligent efforts) at a *minimum* you have worked ~6.8 years on this *single* project (more like 10 years though) and have *nothing* at all to show for it!!! I'd say that time has already given you an answer, but just like the answers you have been given in this newsgroup, you aren't listening! -Pete
From: Bill Snyder on 31 Mar 2010 19:10
On Wed, 31 Mar 2010 18:50:41 -0400, "Pete Delgado" <Peter.Delgado(a)NoSpam.com> wrote: > >"Peter Olcott" <NoSpam(a)OCR4Screen.com> wrote in message >news:6Z6dnbu7zLjhIi7WnZ2dnUVZ_uOdnZ2d(a)giganews.com... >>> The spec would provide the answer that you were looking for as would a >>> search of this thread if memory serves me correctly. Granted, neither >>> method provides the answer in the quick, concise format that you desire >>> but then again it seems to me that the vast majority of your design >>> decisions were made using a "Magic 8 Ball" technique anyway and thus are >>> not based upon anything but fantasy to begin with. Why not cut out the >>> step of posting in this newsgroup and ask your question here? >>> http://web.ics.purdue.edu/~ssanty/cgi-bin/eightball.cgi >>> >>> BTW: I asked the Magic Eight Ball if "Peter Olcott will ever finish his >>> OCR product" and the answer given was "My sources say no." >>> >> >> Time will tell. I have 20,000 hours invested so far. > >Assuming that your 20,000 hours figure is correct, from my calculations, if >you had worked 8 hours a day, 7 days a week, 365 days a year (not counting >leap days which I will give you as a holiday for your dilligent efforts) at >a *minimum* you have worked ~6.8 years on this *single* project (more like >10 years though) and have *nothing* at all to show for it!!! > >I'd say that time has already given you an answer, but just like the answers >you have been given in this newsgroup, you aren't listening! <de-lurk> He didn't say it was 20,000 hours of his *own* time, and a look at the total amount of effort devoted to these threads might explain a lot. Couldn't everybody just take it as established at this point that he'll be revolutionizing OCR at about the same time Arthur T. Murray revolutionizes AI, and move on? <lurk> -- Bill Snyder [This space unintentionally left blank] |