A screed on why I plan to adopt Weblocks and GBBopen... and some plans... It is said, any sufficiently advanced application re-implements half of lisp. Any sufficiently advanced lisp application re-implements much of GBBopen. The original GBB was so far ahead of its time... Take elephant, for instance - almost a re-implementation of a blackboard. There are so many name space and utilities duplicated. GBBopen as an extension to lisp provides you in-memory object management (full MOP plus GBBopen extensions), thread management, sockets, coarse gain scheduling, dynamic re-ordering of execution, and the blackboard for incremental problem solving with multi-lisp, multi-operating system compatibility layers. GBBopen is also the perfect framework to wrapper-in lots of disparate systems written in other languages.
Weblocks provides a nice way to render lisp objects in general via the web. First some background thinking on re-distributing computing: The Web as the UI The web has become the preferred UI for any reasonably complex application - because hyperlinked text is a richer and more powerful way to express and exchange information. The other advantage of the web is the stateless high-latency way of expressing information that is rendered remotely. An advantage? Yes, compared to X-windows, the web wins. It also allows for client-server separation of processing allowing PCs with advanced operating systems to just become glorified VT220s connected via SSL to central "mainframes". See this screed on SSL - http://iang.org/ssl De-centralizing the web - or putting the crypto in people's hands. The SSL web model results in the e-gold and Gmail model. Big brother can index your email and see all your transactions, allege thought and money crimes at will. It also results in huge multi-billion dollar data warehouses for analysing all this. Solutions - In brief, PGP wins over S/MIME, OTR over PGP and ZRTP over SRTP+TLS. In short, the crypto must happen on the PC, and the PC must be secure. Putting the web into the computer Imagine a suggestion that a better, richer experience will come from a higher latency web where all the low latency information is on your computer! Solving this involves solving the tragedy of the commons - or why freenet, tor etc., have failed as technologies. It also involves thinking about persistence, memory management, scheduling, queues... to software distribution plus messaging and money. Moving the web server onto the PC. Content on the server on the PC can just as handily hyperlink to http content on the web if the PC is connected and this is desired. If not, those pages can be retrieved asynchronously. This allows the integration of a whole new set of "web address space" that is on DHT. Solving the ecommerce problem Current day ecommerce is built on a "tell the merchant your secret swiss account number" model. He'll take just enough and send you your stuff. This is layered upon a massive financial ponzi fraud scheme, the USD. See blog for screeds on the fraud, money and what is lawful, and what is meaning of freedom (as in "I live in a free country"). Proposed solutions are described on http://gsf.li/ Getting started Ahem. We need a UI. Here comes Weblocks. UI gotta say hello. Here comes a CMS in weblocks. People need to communicate and trade.... all that to follow. This stuff will have a project website once the CMS is ready, and a dev mailing list. The code will be a Weblocks application, hence this post. On 10/18/09 22:30, eof wrote: > Thank you for your answers. I have read that essay and it was a big > part of what got me really intrigued about lisp in the first place. I > will spend the next few days pouring over Practical Common Lisp which > is sitting here in hard copy on my desk. > > Geoff Go right ahead, you are on the short path to mastery over computers. Cheers, ---Venkat. http://www.rayservers.com/ http://www.global-settlement.org/ > > On Oct 18, 12:37 pm, Rayservers <[email protected]> wrote: >>> It is. You hardly ever want to go back. ;) >>> Weblocks makes you feel you're taking the most effective path for >>> most of them time, and you don't get to repeat yourself often. >> This is true of lisp in general. Our current site is running Portable >> Allegroserve and GBBOpen for a in memory DB. It is moving to Weblocks. >> >> Its faster to write an application in lisp than it is to install and figure >> out >> whats going on in many cases with other languages. >> >> See Paul Graham's famous essay:http://paulgraham.com/avg.htm >> >> We have an ambitious project, we will use weblocks as the web engine for >> rendering GBBopen objects. Source will be released when we have something to >> demonstrate. First step is CMS-like. Back to work on it, I only get a few >> hours >> a week at the moment on this. "Sometime soon", it will have paid staff >> working >> on it. >> >> Cheers, >> >> ---Venkat. > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "weblocks" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/weblocks?hl=en -~----------~----~----~----~------~----~------~--~---
