Saving old software from extinction in the age of cloud computing

Moving our usage totally to the cloud is not a great idea, it seems. Yes, we get convenience, but at the same time lose autonomy, control, and freedom over our computing.

RMS has a good (albeit slightly extreme) view on this: https://www.gnu.org/philosophy/who-does-that-server-really-s...

This view has people discussing the issues e.g. http://arstechnica.com/information-technology/2008/09/why-st...

All in all, if we as users demand web apps that respect our freedom, and employ non-proprietary file formats, then this would be a non-issue. There are a number of ways this could be implemented... one would be to provide a software package (with source code), available to be downloaded and run locally on one's machine, if the product changes or gets trashed. Of course, complete access to your online data is necessary (i.e. ability to download your data as when you please).

An interesting though, especially the view on MMORPGs.

I wonder that he did not name Second Life as a prime example of how cloud based content could rot. First mainy normal products are no longer available, because the creator stopped playing and selling them. I still get weekly PMs about old products, even if I do not play SL for about 4 years. Often about products that are not mine, but where I only contributed, e.g. in megaprims or scripts, and that somewhere show my name. People hope that I have a full permission copy, and could sell them under the hand. Sometimes for extreme prices!

An other example are simulators that go away. Main income for Linden labs is land tax and content just poofs away, when a player decides that paying land tax no longer makes sense. Linden labs will restore the island from backup and run it on own costs, in very few cases like Svarga, if massive player demand it. But I can only name this one case. e.g. 6 years ago more then 3 dozen dance clubs showed my DMX based light and dancefloor system. Today its only one club left. I dont know if my Laser sailing boats are still rezzed some beaches, even if there was a fleet of a few hundred racing regular. And I can think about a good dozen of my products where I'm sure, there they are nowhere displayed anymore.

Last we have classical dependency rot. Second Life allows scripting, and scripts can communicate with other scripts on same and remote simulators, can act as web clients or web servers, xml-rpc clients and servers and mail clients and servers. Some products require a backend to work, so the product will stop working, if the player leaves the game, and stops paying the hosting cost of the server backend.

well there is is OpenSim, OpenGrid and other SL server grid clones. Its nice to run your own grid at home, just for backup of own designs. But you can not import designs from the maingrid, you can not talk to people on maingrid, you basically created a single player massive online game backup.

This article isn't really about old software, it's about software that's current, built on current webapp technology, that will eventually become old. Meanwhile, real "old" software, in use by ten's of thousands of enterprises worldwide, will still be struggling on, nurtured carfully through the developer renaissance, to fit on infrastructure that was never designed to host it's like.

My recent trip to the Computer History Museum (http://www.recursion.org/2014/10/16/a-night-with-the-ibm-140...) had me thinking about this.

I was struck by how _physical_ the exhibits are. For the most part, the exhibits are old hardware. In some cases, they have the hardware running software that you can use.

Software from the PC era can be preserved with emulators (assuming its copy protection can be defeated), but what about internet software? The museum has a _photo_ of Google's first server, but not much about Google's software. Should we be preserving screenshots of Google results so we'll know what they used to look like?

After Google Reader and delicious both died on me, I now try to store all data on my pc.