(By the way, methinks that the "TMSR" bit is somewhat redundant -- it's not like "meanwhile dead ex-TMSR work", or "TSIE work" is anywhere on the horizon.)
This July resulted mostly in Hunchentoot code reviews: taskmasters; requests and replies; and a continuation of that one. So I ate and documented more than 90% of the thing and I'm ready to use it in the battlefield with the reasonable certainty that it won't fall apart at the first prod.
As for the future: at the beginning of this week I had a todo list, and it looked like this. First, the stuff that burns:
- this blog needs a comment mechanism;
- all this Hunchentoot review work means nothing without a genesis;
- #trilema and other castles require a fleet of loggers to keep things running; and
- feedbot has some new patches that need to be merged into the current working tree.
And as for the stuff that isn't (or wasn't) burning (yet I still find important):
- file pastebin;
- there's been some recent discussion of Adalisp, which is still in need of a review and further work to be made usable;
- I want to review older Lisp works, how else to even know if something "better" is possible?
- eventually I'll want to steal and review all actual and potential Lisp infrastructure;
- I still haven't genesized The Tar Pit; by the way, is there anyone else out there interested in using it?
- I still haven't run a Cuntoo to this day, y'know;
- there's a ton of other work there; I'm not sure how much of this I'm going to finish during this lifetime, so at some point I'll just sit down and re-prioritize.
I also had a plan for the following four or so weeks -- in fact more like the next eight, but plans tend to change along the way. Anyway, the short-term plan, word for word, was:
- Week 36: genesize Hunchentoot and everything else required to run it, save for the operating system.
- Weeks 37-38: implement The Tar Pit comments and publish said coad1.
- Weeks 39-40: read Stan's logger; stand up a Logbot; write a minimal WWW frontend that takes items from the logotron database and renders them; publish that.
Then discussion happened, and... well. I'm maintaining a Lisp blog, so I guess I might as well add comments to it, since I have a Hunchentoot genesis almost published, and since the painstaking labour involved in moving every post to MP-WP (plus fixing the links) would take a yet-to-be-determined amount of time. For the same reasons, I'm guessing I could take a stab at a logger WWW front-end based on logbot... or should I, really?
I don't know, to be honest. For the record, I very much agree with Trinque's comment: I'm skeptical that the "Lisp environment" can deliver more than the "Python environment", and it might yet deliver less than the "LAMP environment" and in the end prove to be a monumental waste of time. I'm not happy about the prospects of getting stuck in a solipsistic hell of my own making, so really, the thing that mainly motivates me to continue working on this Lisp stuff is other people finding it useful. Is there anyone else in the house who finds any use for this?
I also agree with Alf's comment, there's tons of (burning) problems out there to solve, not like I need to add extras on my plate. So then I'm bringing this before the wisdom of the forum: what do?
-
I actually have this broken down into pieces, let's look at it.
First, the posts need to be indexed somehow in some kind of database, so that the blogotron knows how to retrieve the content and the comments for it and render everything into a page. I don't care about the database backend, the code that does this should be attachable to any type you like with minimal glue.
Then, I'll need to make some modifications to the blog itself, so that it knows that such thing as "comments" exist and can be retrieved from said database and rendered. This includes page layout modifications as well as backend scaffolding for getting comments and inserting them below the post.
Finally, there's supposed to be a way to post comments, à la Wordpress' wp-comments-post.php. Additionally, this should come with anti-spam support built-in, but... well, did you see the "stuff that's burning" list? I'll patch that wound when (yeah, when, not if) it starts burning. Until then, I guess all comments are to go to the moderation list, or I'll whitelist IPs or whatever.
Oh, and by the way, I should also implement server-side selection for this thing at some point. ↩
[...] plan last month looked very good on paper, and yet reality went quite differently. Summarizing the CL-WWWism work, I've come to the [...]
[...] Other than that, there is no "Hunchentoot versus Apache" contest to speak of, no matter how you choose to compare the two. So if I set out to write a Common Lisp web server from scratch, given my experience so far, I'm not sure I'd do a much better job than Hunchentoot (let alone Apache), at least not unless I spent a year or so (though I'm afraid this could easily turn into "a decade or so") reviewing WWW servers in more depth. And so, the question comes, is this really worth doing under the circumstances? [...]