Plugin crashed

I’m so sorry! Thanks for bringing this to my attention.

I’m still (re-)getting used to the version control system (SVN) that the WordPress plugin repository uses. It has been years since I used it!

This should be fixed now if you download again. I have not promoted the version number.

Joe.

So. Plugin installed. And…

“Warning: Division by zero in /var/www/admin/www/MYSITE.com/wp-content/plugins/iterative-headlines/headlines_calculator.php on line 34”

Im your patient testing hamster. )))
But I like your idea so much, I’m ready to continue.

I appreciate your patience. Please update to the latest version (1.4) where I have moved everyone to the latest version of the API. I’ve done extensive testing on this version — including on fresh installations of WordPress — so hopefully all the issues you’ve had thus far are gone. ??

Joe.

Victory! Works good.

A few notes. I’m not boring. Just want to make perfect.

1) Tip above headline field: “Your blog appears to be written in Russian. We don’t have much advice for non-English headlines at the moment. Don’t worry, the artificial intelligence still performs well in your language!”

Right? Not depends of language?

2) In menu not all buttons has hover-effect. See screenshots:

https://i.imgur.com/HmdJ00A.png

3) I mean I need a text under the “worse” label”
https://i.imgur.com/aWSHJAD.png

like under “better”:
https://i.imgur.com/cBLCCF3.png

4) The plugin is able to move in time?

https://i.imgur.com/aUXIu1w.png

Or he considers me? Its not good, because statistic will not correct. May be no need to take into account the administrator?

5) It would be useful to have at least a simple statistics in the administrative panel on how to operate the headlines. I have a few ideas about design, but I do not know whether it is possible to implement them in your code.

This is great feedback. The plugin currently takes in to account the administrator’s behavior (as you are seeing in #4), but that is a bug, I will fix it shortly. The other advice is all great.

The basic version of the plugin is not language dependent (it works fine in any language) – we do not currently consider the content of the headlines. The premium version of the plugin considers the content of the headlines and other properties of the user for selecting the best per-user approach.

If you get a chance, can you elaborate on point #5? What do you mean statistics on how to operate the headlines?

Joe.

А new piece of nagging. )))

Look at states of loading, please:
https://i.imgur.com/ZsXsDxN.png

*yellow is waiting, blue is connecting.

And total time:
https://i.imgur.com/JtjxRLF.png

I made five attempts. The same result.

I repeated that same test (pingdom) a few times this morning and cannot reproduce the load time you see. I am getting around 50ms negotiation/DNS/connection and 2ms content time, for a consistent total of less than 55ms load time, both locally in Canada and from Pingdom out of NYC. This is going to be dependent on your users’ location and network connection, but continue reading:

The log and success files are loaded at the very bottom of the page, so they should not effect the user perception of speed — the page will render just as fast as if they weren’t there. Those requests are made in parallel and at the end of the run and in most renderings won’t even start to be requested until the page is starting to render.

That said, my server hosting this stuff is in California. From my ISP in Canada, I can achieve 50ms round trip requests to it. If you are across the ocean, there may be some structural delay that will necessitate a server somewhere in Europe. I cannot set that up today, but I will do some testing to see if a significant issue exists and if it warrants adding a second server for geographic diversity.

Joe.

Yes, I checked your guess. You’re right. My site in Europe. When I test through the European servers, the site is rendered fast, your server responds slowly. When I test through the American server – is the opposite.

I made in the Google analytics report about requests and compare it with the plugin work. While 50% of the match. I continue to watch.

If you get a chance, can you elaborate on point #5? What do you mean statistics on how to operate the headlines?

I mean, you have to show the summary statistics of all the titles and everyone individually. I have ideas for the design, but I do not know whether it is possible to implement in the code and whether you have such plans.

I will see what I can do with regard to deploying multiple servers (one in Europe, perhaps) before I expand the pro version to public access. I appreciate your testing. ??

Joe.