This is a project that in this moment is under development. I am the only developer and it is a very practical utility for PaBa network, a Internet web sites network founded and managed by me and Berce Alin, an old collegue. 1st what is PaBa; this is a network of web sites that contains a lot of sites with diferent services: Internet services (proxy, geolocation, band speed), travel and fun(hotels, travel blogs), Internet Imoral Utilities (this are from Boby - Yahoo Status, HI5 acount), Cultural and religious web sites (Historical personalities, churches, religion, history) and more will come from now on. Also we accept and other memebers in our network but we have to define some conditionsfor accesing and using our network, specially the performance and the quality of the candidate ofered services. So this is a big network. Will be, I hope that this will happen next week.
Big network require big chalenges. This means that we have to link this web sites into a network; this will happen using a page on each site, where will be link to each other sites from the network. So I try a lot of "simple" way to do this. We try to use a common database that should be accesed from all the sites; no way, one our hosting providers has blocked external requestest, of any kind (database request, http request, ftp request). That means that I cann't use XML Web Services, external XML web files, external requests of any kind. Other simple solution that was aborted afer one second from it aparition was to include this links into the client side; no way, very bad ideea: this link should help the network to increase. If this is only ar client side (using javascript, iframe or frame or other ways to include external content, on client side) this doesn't help the network specially when we speak about search engines. It should be generated by each site.
So what I sugest: using an xml file that will be uploaded on each site, over the ftp, helped by xslt technologies we generate on server side this pages. In this way the search engine will recognize our network and the users browser will get from each site server the list with the links, ordered by relevance.
For this, I create an ftp robot program; this means that a list of ftp connection settings (server, users, password, directory) will be stored somewhere in robot settings file; I keep the sites in a special database from where I generate the XML files. My robot takes this XML files and send this files to HTTP servers. From here each site publish the XML transformed using XSLT in HTML format and we get the wished result. Now I have to find something that can stop us (specially from hosting providers).
In this moment the ftp robot is under development (more exactly it is developed 30%). When it will be functional, even if is not finished, it will be published and anyone can download it. Untill then please wait (this should be finished in less than a week).
Thanks. I wait for your opinion.