Error message from Firefox

Recently I often get this screen when trying to return to New Posts from a thread page.

Network Protocol Error

An error occurred during a connection to photography-on-the.net.

The page you are trying to view cannot be shown because an error in the network protocol was detected.

Please contact the website owners to inform them of this problem.

This is most often caused by cache or a cookie being corrupted on your system.

You can first try reloading -- hold Shift down while pressing the reload button upper left in the toolbar. If that fails, you may need to clear the cookies for POTN.

Thank you. Reloading is Command-R here, and I've tried that. Sometimes, in addition, loading a POTN page takes so long that I press ESC and try again. This usually works.

Would deleting POTN cookies lose all my preferences for forum filtering, margin width, and so forth?

Whenever things get squirrely, I always clear cookies and cache. That usually does the trick.

Yes, do that ^^^

OhLook wrote in post #18811032
Would deleting POTN cookies lose all my preferences for forum filtering, margin width, and so forth?

No. That is stored on the POTN server.

Also make sure FF is fully up to date

Well, if the problem is at my end, why does Firefox tell me to inform the website owners?!

OhLook wrote in post #18811091
Well, if the problem is at my end, why does Firefox tell me to inform the website owners?!

It may not know it's at its end. Probably figures out whatever is cached and whatever cookies it has are all good. No way for FF to check.

From what I can tell, Pekka has some pretty advanced cache things going on. Performance on this site is really, really, good. By pushing the envelope it may create some confusion on the browser's end.

Log out, clear your cookies and cache, make sure your browser is up to date, log back in.

If you still have problems, then it might be something going on here.

BTW, there is nothing potentially bad about clearing cookies and cache. Most people, myself included, do it from time to time. The only potential problem is if you have a cookie set for a website that you don't know the password for ... you might have to jump through some hoops to get it back

gjl711 wrote in post #18811099
It may not know it's at its end. Probably figures out whatever is cached and whatever cookies it has are all good. No way for FF to check.

Bingo.

OhLook wrote in post #18810450
Recently I often get this screen when trying to return to New Posts from a thread page.

Network Protocol Error

An error occurred during a connection to photography-on-the.net.

The page you are trying to view cannot be shown because an error in the network protocol was detected.

Please contact the website owners to inform them of this problem.

I'm pretty sure it has nothing to do with cookies.

This error is probably given because your ISP (or VPN) cache's pages - or https negotiation failed for some (or same) reason. The message is not really useful for any debugging, sorry.

BTW: we do not serve cached pages ("compiled" pages that are stored for re-use after an initial request). All pages are "built real time". The response time is fast because the code and server is fast

Do you not cache anything special locally ... on the user's machine?

I tronz almost exclusively on my phone, and don't have the best signal and don't use wifi. POTN is head and shoulders over anything else.

Actually, as I'm about to hit submit, that is one of the things that is so quick. Most other sites will reload the page when submitting something.

Left Handed Brisket wrote in post #18811274
Do you not cache anything special locally ... on the user's machine?

Only what the browser caches, cookies, javascript files and html. There is nothing else in browser storage areas. AMASS is designed to transfer only what is necessary on a subsequent page load, usually 10-50KB (without images). That's fast with even a bad mobile connection. When you submit a post, nothing but the result response is loaded.

Left Handed Brisket wrote in post #18811274
Actually, as I'm about to hit submit, that is one of the things that is so quick. Most other sites will reload the page when submitting something.

AMASS uses AJAX, original coding, unique database design, Redis, JQuery etc.

PS. There is a lot of talk in Finland (maybe elsewhere too) that there is a serious lack of coders. Well, because they will not even consider anyone over 35, and require latest "popular" frameworks to be used, they will miss a lot. Coders seem to be more like "robot slave labor", than creative problem solvers and logically thinking people. Just my €0.2.

Kind of the same over here in the US.

I built my first website in 1994 or 95. I did mostly print design, color correction for print, etc, but knew www was going to be huge. I struggled to keep up with both and ultimately the print and photo side won the battle. I stopped worrying about Action Script, PHP, whatever by 2000 and concentrated on design while hiring out the stuff smart people knew best.

I know a couple 35+ guys who went back to school and within the last 5 years or so entered the industry with good results, but I can see that not always being the case.

I also agree that there are many executives/managers in the industry that just want the job done whether it is pretty or not. People don't want to take the time to understand that there is nuance to producing something great, great code or great photography for example. I worked as a photographer for a local manufacturer but was surrounded by coders and engineers. They did a great job with project management and execution of some pretty complex stuff. However the coders and I both reported to the same sales/marketing idiots who wouldn't take the time to listen to us about the unnecessary and arbitrary constraints they were enforcing. The end product suffered tremendously and would require more work in the future because concerns weren't heard.

Anyway, I hope at some point executives will have enough knowledge to let the project managers and engineers do their thing without interference. This can work with either simple acceptance they don't know wtf they are talking about, or enough peripheral knowledge that they can add a little to the project to be of benefit. We aren't there though, and I think that is leading to the belief that coders should just be cubicle monkeys cranking out their code.

Don't know if that makes sense, but that's my story and I'm sticking to it.

lol