Mail Servers, eCommerce servers, and RDBMSs are obsolete


Webservers, Mail servers, and RDBMS-based business systems and ecommerce are obsolete on grounds of cost, complexity, performance and security. Their functionality can be replaced by a pure, flat, public filespace which the participants would post messages and transactions encrypted, but publicly visible. 


Access to these messages would be controlled totally by the users with no reliance on sysadmins or user-level security, by managing encryption keys. 

The need for transactions that span company boundaries breaks server- based control of security. The nearly universal practice until the last 5 years was simply keeping internal systems unconnected from internet. Security has *never* been adequate to enable global internet commerce-- not in past, not today, and quite possibly not tomorrow. All server-based security is based on encryption anyway. Why not keep the core technology, and ditch the obsolete applications?


Consider that document-based commerce based on standard XML vocabularies enables asynchronous interactions and integration without hubs. ebXML is one example, enabling rich transaction capabilities based on simple exchange of XML (text) files.

Within the global filesystem, pure information would be stored, in standard XML vocabularies. Access constraints would be minimized; a user could select a new Best of Breed provider for various business services by simply providing indexes and encryption keys he controls, without begging permissions and paying fees for additional logins to complex, centrallized hosts and banks.


Hosted filesystems exist today which are utterly, totally robust and reliable and online 100% of the time --and at very low cost. If these storages were available 20 years ago, SMTP, POP3, and HTTP would never have been invented, nor SQL RDBMSs, in their present forms. Certainly, their markets would have been much smaller.

Messaging, transaction execution, and accounting can best be done on shared filesystems with minimal server logic. The problem of security and data standards so great, for small /medium businesses, that it begins to dominate their whole equation in selecting internet commerce models.


Imagine if Exodus or other ASP provided a new, global file service. It could be open to the whole wide world, or members only. It could be one of the new anonymous filesystems (Freenet, Publius, or Mojonation) or a more traditional SAN or other file space. 

Let anybody in the universe create files or directories. Make the service writable but then read-only, and all files globally visible. The ASP might limit sizes, charge for uploads, or expire all content over 90 days, etc. Or maybe, require user to specify file expiration date upon each upload, and charge micropayments accordingly.

The DNA would organize itself to take advantage of this new petri dish because it would be cheaper and more practical than complex server logic.

The storage logic of applications would have to be re-written to continually create new files, rather than edit and delete files. Data storage would be in smaller chunks. XML is appropriate technology for this. 

We could publish basic web content such as HTML files up in the sky. Rich content could be distributed. Nobody owns the commons. Files never get deleted. Nobody can break them, edit or delete them, and nobody can read anything unless you give them the key. All unix security is based on encryption. So, why do we need sysadmins?

We can setup directories and call them inboxes. We could send messages, B2B XML transactions and other content to each other by encrypting them with PKI and putting them into each other' directories. E-commerce is enabled by years of hard work in uncoupled, document-based commerce.

We could form shared transactions and collaborations in the open filespaces. Legacy web developers will be aced by these complex XML structures, quasi-executables up in the sky. The chains of backward compatibility would be loosened.

We could conduct ecommerce based entirely on exchange of encrypted files. We will create robust slow OLTP systems by pushing new versions of records to a shared filespace. 

Putting conventional application files on the server, as well, nobody would EVER be unable to get to their data. You could map a drive up in the sky, with encryption hidden from the user or the application.

Individuals and SMBs wouldn't need an RDBMS and wouldn't need a webserver, mailserver, FTP server or other hosts which require you to trust a system administrator. Most security problems derive from this original sin, of trusting an intermediary. And you just get FED UP with all the data formats-- you just want to store pure information, messages, documents, and transactions, in XML.

Hub-and-spoke transaction architectures with human-supervised security are a cultural artifact. If you trust the individual and the free marketplace, you would build a common filespace with intelligent nodes and a common vocabulary. But if you own the planet and you want to control the nodes, you will build giant RDBMSs, terabyte data warehouses, B2B exchanges and webserver farms.

Todd Boyle CPA Kirkland WA  425-827-3107

* If it doesn't have prohibited content, then, apparently, it's not secure.
* If the government doesn't want to ban it, it's not secure.
* If the government wants to ban it, then, perhaps, it's secure.


Peer to peer workgroup
Sharing is a Virtue