Rondak's Oracle

News and fun things about the games we play

Category: Technical (Page 1 of 3)

Site News

Not much new. A few more lines of code, still plugging away as I can. If anyone here has a better suggestion for a C++ –> WebAssembly + JS compiler than Cheerp, I am all ears. I’ll play with emScripten + binaryen, in the hopes that will work better for me, but maybe someone has an idea?

Site News

Pathfinder playtest started; Thorne has assorted channels and forum posts to organize, if anyone is interested. My impression so far is mixed – the changes look interesting, but they are going to take a LOT of getting used to.

In site news, I have been making good progress on the code and am almost ready to put a test page for the authentication and account management system up. Might not make “Summer” as intended, but I am making progress.

Site News

Still working on the new codebase. About halfway through getting the auth system to behave and integrate with OpenID providers; starting to give thought to how to organize actual site functionality.

As a PSA, I would continue to recommend against shared-server (cloud hosting) arrangements, due to the prevalence of side channel attacks on modern CPUs. There’s, of course, the well-known Spectre class of attacks; a similar principle can be used to leak encryption keys when using a modern Intel or AMD CPU’s ability to perform AES crypto in hardware; below that layer, modern CPUs have what is called a “management engine” to perform system initialization and maintenance functions. As it turns out, the firmware powering these MEs is also vulnerable to attacks, including remote-exploitable arbitrary code execution attacks. All this leads me to believe that the course of wisdom in hosting arrangements is, at minimum, dedicated servers. Ideally, dedicated servers on a dedicated network, but I am unfortunately not aware of any providers offering that, shy of renting a rack, deploying one’s own hardware, and one’s own firewall.

Site News

Finally had the cycles again to work on the new codebase. Taking a break, while ${WORK} was a pain, did help me become clearer on some more unexplained bits on how to implement an OpenID identity provider, so that’s good. Next, actually write the code that gives life to the flowchart I built this weekend.

Site News

No real progress on the new code. ${WORK} is killing me. Rondak’s Portal has acquired a Discord server; this should be considered temporary for now – while voice chat is on my feature list for the new codebase, I have not yet decided whether to integrate Discord or opt for another system.

Site News

Not much to report, was mostly pulled away to deal with paid ${WORK} this week. Still, I got a little bit of work on the new Portal code done, and squashed a few more bugs in the existing code. Progress. Slow, but progress.

Site News

Making progress on the new codebase, slowly but surely. I settled on a backend interface to actually run the Portal codebase, and got a toy application running on the framework to see how parameters are passed and what I need to do to make the code work. Now in the process of getting it all integrated for that pre-alpha site I mentioned I wanted to have up in August.

Site News

Some more progress with the new codebase. Got most of the HTTP protocol handling dealt with, I need to extend the POST message body parser to also handle JSON objects. That should make writing the actual server code that much easier, since I won’t have to figure out how to send all my headers by hand for every page.

Site News

Not much new to report this week, I spent most of it working on things that actually pay the bills. That having been said, I had some time to at least think about Portal development, and talk to Sylverthorne about getting started on mocking up the actual login flow – as the supporting code for that is going to be complete soon and I will then start building that (very low-functionality) demonstration / proof-of-concept site that I set myself a Summer 2018 deadline for.

Site News

Just a quick status report. I’m almost done with the initial implementation of the authentication system for the new codebase; the server side is written, compiles and passes basic functionality testing. While doing the work, I noticed a few gaps in my underlying libraries, so I will be going back to infrastructure work for a while to fill those gaps and factor out reusable code from the auth server component into the general infra library, then I’m getting started on writing the client side of the auth code.

Page 1 of 3

Powered by WordPress & Theme by Anders Norén