> On Jul 25, 2022, at 3:21 PM, Glauco <[email protected]> wrote: > Impressive job, David.
Thanks! Here are a few additional notes. I'm slowly working on getting the "rebuild metamath website scripts" working on a new system. One problem I've found is that the current scripts assume that the website is already working & downloads from it. So if anything fails, there's no way to recover things. That sounds like a bad idea. I intend to change things so that *all* bits used are: 1. From a public source (like Debian), 2. Version controlled publicly (e.g., GitHub), or 3. Generated from scripts that are publicly version controlled & only use data meeting criteria 1-3. That way, the website is *only* generated from publicly-available information (other than private keys). It may be important to make a few more repos & split things off, e.g., maybe symbols should be in their own repo. That would make it easier to do that, and the results should be easier to manage. There are some old files from Mel O'Cat, including a whole bunch in <./ocat/mmj2/$RECYCLE.BIN/>. I doubt we need them any more (if we ever did). I'd like to eventually put us.metamath.org behind a content delivery network (CDN), probably CloudFlare. This means that requestors will request pages through a CDN, and if they get a page, the CDN will keep a *local* copy near the requestor. This makes requests faster, and also protects us from distributed denial-of-service attacks. These could hurt my wallet, since I get charged for requests more than a certain amount. --- David A. Wheeler -- You received this message because you are subscribed to the Google Groups "Metamath" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/metamath/1B3D768E-20C5-41AE-A791-55A488B6C451%40dwheeler.com.
