yes, spambots are a real problem and after all this saturating work done, the issue is -still- not solved, next thing to try: "do not include the form in the html, but generate it from JS or popup it from a button, in order to avoid bots finding the forms in the page"
now, the biggest biggest issue is... how to remove all the collected "fake emails" in the databases? I can process the existing filters made (already doing this since weeks) but stills, many emails are "real ones" (in other words: unsolicited email added, so real emails inserted by spambots), so its not possible to detect them since they are real... the only possibility is:
to send a future "confirmation required" email asking the people to re-confirm their news subscription, in a way like "sometimes we need to cleanup our database of emails, if you still interested in elive, confirm your subscription", but of course, the bad-side of this is that we will lose A LOT of subscribers by doing that.. (many people won't see the email, many won't care enough -maybe this is fine-, maybe they never received the email due to spam filters, maybe they missed, maybe they didn't understood the message), so, we will lose a lot of them from that
also, spambots are really smart today, they even click in the links on the received emails like if they were humans
the most illogical part of this is that spambots can use javascript too. archiveteam (an organisation that aggressively archives websites) actively used a bot named "chromebot" which could get around cloudflare and use javascript before it was decomissioned.
its a very delicated thing since is so hard to avoid spambots (and thus, allowing spambots means to have fake / unsolicited emails in the database which ends in being blocked / marked as spammers in email providers)
Thanks for the tips, today I will do tests with the JS-based form and these cases, but note that I think today's is almost useless to have javascript disabled (even google recaptcha requires js, i don't think it works without it?)
google recapcha v3 is enabled now, which is not interactive and based in a user score
yeah I think its easy to make bots use javascript too (this makes 100% sense making a bot), but I will still do a few tests and see how it works (to at least see if bots are actually able to see / use the form, or they are skipped entirely)
btw there's already 20 IP's which used the honeypot and turns to be banned for 6 months, most of them are from russia
The FAQs page is entirely rewritten, so I have get rid of an annoying "faqs" plugin and converted all the faqs to individual posts, the result should be the same especially for SEO (so should not have a negative impact), the website should improve in stability but especially should be more light/fast now (probably unnapreciable)
But, don't do it yet, I will try to create a temporal copy of the website where everything can be edited (in fact, only these ones for now, since all the contents are going to be rewritten), so its much better, easier and faster to edit directly the contents on the website
Wouldn't it be easier to do that on github or gitlab ..... or even here as a wiki post?
That way we'll at least have a way to backtrack if we deem a change to be undesired or wrong.
Then again we are so few that it might be overkill.
Good idea to have stuff split into smaller chunks. Makes it much easier to oversee.
With the whole site I wouldn't even know where to begin ...... that's an awfully big chunk of text.
not really, the editor in the website is so handy that is more fast to edit directly on it
about backtracking, since its a "temporal website" is not a big issue since can be verified before to become public, by other side we have also "revisions" in the website (on which you can recover a version of a page from X days before), the only issue is that we cannot have a real visual "diff" of what changed, except, by creating a small cli tool that fetch the contents (so, words) "before and after" which I think should not be hard to do
yeah I try to do it in the best "less work" as possible
I suggest changing the "Download torrent" on the old versions site to the archive.org listing or at least link to the archive.org torrent (not sure if it's the .torrent or _archive.torrent) ... too busy to do the myself.
My name is Samuel F. Baggen and I'm the founder of Elive since 2002 when it was a very different, geekier system, since then I have been the solely developer of Elive and received extra collaborations by volunteers, later, the first public version appeared on 2005.
That is a run-on sentence. The best way to fix them is to just split them up. Example:
My name is Samuel F. Baggen and I'm the founder of Elive since 2002 when it was a very different, geekier system; since then I have been the sole developer of Elive and received extra collaborations by volunteers. Later, the first public version appeared on 2005.
Hummm, that's the original sentence I remember correcting for exactly that.
It became (on webtest2 i.e not published yet):
My name is Samuel F. Baggen and I'm the founder of Elive, back in 2002 when it was a very different, geekier system. I have been the sole developer of Elive since, occasionally receiving collaboration by volunteers. The first public version appeared in 2005.
But the example is clear: Keep it simple and short!
Considering how long these corrections have been available, it is indeed high time @Thanatermesis updates the text portions on the website accordingly.
I don't remember.. it was there an improvement on that torrent?
Actually to have the file on the server has two advantages:
full uptime & faster download
downloads of the torrents are tracked in the stats website, this makes us to know how many downloads the torrent file got (I just configured some hours ago the stats to be able to count it better)
thanks for the tip, I think they are updated now
thx, updated
oh, yeah, the webstest2 site was not yet integrated on the old one, I don't remember why this task was paused due to some reason (but the FAQs page was imported I think, I still need to review/work on that)
yeah agree
in fact the ones to edit should be made on webtest and not the normal website (on which you* have access), but we have not a control/workflow on that at the moment so let's wait for the future work on the website
note: I'm working now on migrating the server, based on elive-for-servers, new email server will be ready soon too!
Oh i see, so i just need to use the new .torrent file and it will be seeded by 'internet archive' (among others), right? - do you have the link around?
The server of Elive (main elivecd.org website and other extra services like keyra) is officially moved to VULTR !
Changes:
We can already notice a small speed increase
Distro base is bullseye
Email server is set up new and we should be able to send back to hotmail addresses for example (if you have one, try the downloads process to see if you receive it good! and mark is as its not spam)
next: the forum server too, which also has hotmail sending issues...