Posted: Fri Jul 10, 2020 6:26 pm
Although great ideas, this is a non-trivial operation. The code that skyscript runs on is an older version of PHP and translations to newer versions of a coding language often means that necessary functions and operations become deprecated - that is, no longer work. Additionally some newer server configurations only support newer versions of PHP.Michael Sternbach wrote: I would be all for updating the forum - and not only in regards to potential log-in issues. Modern forum software allows cool things like direct embedding of images (a number of my earlier posts have become more or less useless because the server that once hosted my images went down!) and, hey, even videos!
This means to update the server stuff might mean updating PHP and that means that, realistically, many things will break. The scope of that breakage is difficult to ascertain as there are thousands of lines of code of PHP.
Another approach, a preferred one in my opinion, would be to move to a new server with new code, fully accepting that Skyscript as it stands now will not work. What does that mean for, not just the forum, but the rest of the site? Realistically it's possible that some pages won't work and others will, or, that none of the site works. This would be very difficult to guess or ascertain. In order to account for it you have to basically recreate the site and all of the files which make their way into that site. At last count there are over two thousand files which account for the site, not all of them are 'content' some of them are configuration files and 'code' files which allow for the rest of the site to run. This is not an atypical set up.
So, let's imagine we move everything to a new server and expect it all to break. What's next? Ideally we would at the very least want the forum to stay. Here we run into an additional problem. Deb made the decision when building the site to allow for 'custom' modifications to the codebase - that is, manual alterations of the 'out of the box' code that was used. Those decisions are not documented anywhere.
So let's say we want the forum content, but acknowledge that some features may not be there. We could export the database which drives the content, but then we would need to find a way to manually 'link' up the new server's database to the old one - the easiest way to do this would be to write a series of database related code (likely in SQL or something similar), which would convert the database export into a format that the new server would recognise - in other words translate the old into the new.
Without knowing what the new looks like, that's a difficult project to size and scope. This is doubly so because, currently, I have not fully examined and documented the full database structure as it currently stands. There are a couple of dozen tables, and I'm not entirely clear what they're all used for - some of them have effect only for the administration panel to block and ban people for example.
So, whilst it's a laudable, and obvious, goal to update the site, as soon as we start thinking about it we start quickly running into problems. Just to contextualise this, the 'admin' team of skyscript currently consists of me, and someone who does 'caretaking' on the server. I'm a web developer, so my skillset is well suited to managing and resolving the problem, but I work full time and see astrological clients part time. My free time after those commitments are also taken up with doing a part-time Master's degree course on top of working full time.
At the moment I am not sure how much work it would be to make this change, and what we don't want to do is start the process and be unable to finish it. We also do not have a testing environment to begin making a copy of the kind of changes we'd like to see - I'd have to do it on my own machine, and make sure that nothing links to the live server.
It's worth pointing out, because I know people sometimes think otherwise, that being a moderator or being an admin is a labour of love only - there is no payment and the site does not make profit.
So, in short, I would love to make a whole load of changes to the site. At the same time, people talk about just closing the forum or archiving it. I don't want to put in that work (which could take over a year) only for it to go to waste.
Now, if I can work on this it would realistically be starting at some point in one year from now - that is when I should be finishing my MA, and realistically I do not have time in my life to dedicate to both.
But, my previous degree, and MSc Computer Science, involved working on astrological software to calculate charts with over a dozen house systems and two sidereal ayanamshas to a degree of accuracy of within 1º (I did not use astrodienst, but programmed the celestial motion without APIs such as that). But the point is that if we were to do a rewrite, I would want to include an additional featureset - the ability to post a number of charts alongside every post, one in which the planets, points, house systems and zodiac could be dynamically changed. So the poster might post a horary chart in Tropical and using Regiomontanus, and you, as a user, might choose to change the house system to Placidus or Whole and then hide the outer planets and add the Lot of Fortune etc. - whilst able to 'reset' back to what the original poster selected.
So, I'm writing all this to hopefully contextualise the discussion and raise awareness of the kind of problems that one would face in updating the site. I do, however, really want to make that happen. I just want to be realistic about the ability to do it and the amount of work it might take.
Some of the other smaller problems might be achievable without such a drastic change. I'm currently in the process of completing some research into house systems and the choices that people make when choosing one, or the concepts they employ when thinking about them. I expect that to be finished by the end of August at which point I will try to prioritise any 'cleanup' tasks for skyscript.
I know that it may seem like the moderators or admin (i.e., me) don't care about the site or are just too lazy to do the simple things it takes to improve it, but hopefully some of this demonstrates that it's really a much more complex problem than it might seem at first.
Right now my main concern and priority is just to keep the site running.
Blame it on my blatant Sagi Sun!