*impressed blink*

@Chris: it's the ones you don't see coming that you've got to admire, don't you...?

:))

edited to add signature. And something else I'll get back to.

It was about the text colour choices on offer not having changed, and not being displayed against the dark background they'll be displayed on - but I'll need to organise the sentence more before I pop it up at seti. I also don't know whether it's too trivial to mention there. I won't let that stop me though... ;) or whether it has been already, in which case I won't

this edit is because I thought you might like a bit of punctuation in my last paragraph. You're welcome, except it's: now redundant
:)

---

What I meant was this London webcam

https://en.wikipedia.org/wiki/Real_Time_Streaming_Protocol

It is like http, but different.

---

The one I used used a simple one line .CSV file which was updated at regular intervals by the weather station. The web page read this file whenever there was a page request, using PHP, and generated the HTML for display. I a similar manner the archive data was in a .CSV file that was sent up to the server every hour, and again, PHP was used to generate the HTML file for display.
I can't see why much the same process couldn't be employed for the SSP as that doesn't need proper real-time data, run a cron job to generate the .CSV data file from the servers, and a very simple PHP script to generate HTML for the displayed page - the latter could be controlled by the same cron job, or on demand.

The one I used used a simple one line .CSV file which was updated at regular intervals by the weather station. The web page read this file whenever there was a page request, using PHP, and generated the HTML for display. I a similar manner the archive data was in a .CSV file that was sent up to the server every hour, and again, PHP was used to generate the HTML file for display.
I can't see why much the same process couldn't be employed for the SSP as that doesn't need proper real-time data, run a cron job to generate the .CSV data file from the servers, and a very simple PHP script to generate HTML for the displayed page - the latter could be controlled by the same cron job, or on demand.

Precisely. As the SSP data is also available in XML, I'm sure that is what the cron job outputs.

---

I doubt Seti will change things, not enough people are complaining. only really a handful of people out of many thousands regularly use the boards. They know that.

Dr. Anderson has obviously read and responded to at least a few comments made.
Please do not cut him short.
He WILL and does, change things to suit his agenda, whatever that may be.

But he is not going to just buckle over to criticism.
You must know that by now.

Meow.

With the spirit of Christmas in mind I'll allow solaris (aka speedman/musicplayer/finnenummer/happyday/bluestar/pinkpanther/andprobablyseveralothers) to be here, as long as normal conversation can go on and the delirious postings go into his 'Still here' thread. Go over too many times and this account joins the others at a forever banishment, Christmas or not.

---

Just posted my new FAQs message at the Einstein forums, but wasn't easy to get there. See, I'd lost my password, so I thought to log in with my authenticator.
Did that, got onto a screen where you can change your password, have to fill that in twice and then for good measure you have to fill in your old password to validate that change.... see where this isn't going to work?

And yes, I tried to fill in my new password tp validate: didn't work.
I tried filling in no password to validate: didn't work.

What use is it to log in with your authenticator key, to then be only capable of changing the lost password with the password you lost?
Shakes head. Mumbles something.

---

Any project willing to invest and show commitment to DC and boinc should be encouraged. I've been through a number of website changes and i don't think it's a mess at all, a few bugs -yes, a few features missing, however it is up and running. Changing anything to a new format takes a lot of time, and lets be honest, boinc is withering on the vine of neglect.

---

...and lets be honest, boinc is withering on the vine of neglect.

https://github.com/BOINC/boinc/issues?q=is%3Aopen+is%3Aissue shows otherwise as do https://github.com/BOINC/boinc/pulls and https://github.com/BOINC/boinc/commits/master. With BOINC being developed by the community it is now more productively being developed than it has been in the years there were just three paid developers plus a handful of volunteers. That this doesn't translate into a new client release every week is just because it isn't that easy to do so securely, and no one has stepped up to the plate yet and said they're going to do this.

Back to Einstein being a bad project for new users.

I was looking at their Join Now section, and it's abysmal.
The new user section https://einsteinathome.org/join has 3 options... well, it has 3 points, but after point two there's nothing.
It's as if the maker would come back to it at a later time and never again did, doesn't fricking care.

But it gets worse, for when you click on "I'm a BOINC user" you get the following:

If you're already running BOINC, select Add Project.

If you're running a command-line version of BOINC, first use the BOINC Manager software (elsewhere if necessary) to create an an account at this project, then use that account to connect with the command-line version.

If you're running a pre-5.0 version of BOINC, please upgrade to a more recent version to create an account at this project.

So what to do after the Add Project? Not even telling how to get to Add Project, people will have to guess?
If you're running a command-line version of BOINC... you usually do that because you cannot run BOINC Manager. No explanation on how to use BOINCCMD in that case. And a pre-5 version of BOINC? How old is this text?

The most user friendly option, to register via the website, isn't even there. There's also no user search. There's no option to figure out if a user has already got an account, other than guess by filling in email addresses via the request new password page.

Just preserving a set of error messages I saw at SETI this morning, in case anyone wants to do any debugging.

[As of 3 Jun 2017, 8:40:05 UTC]

Server status

Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/mysql_connect_test oscar' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 326 
Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/mysql_connect_test carolyn' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 334 
Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/paddym_connect_test' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 342 
Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/marvin_connect_test' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 350 
Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/apache_running_on_host synergy' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 369 
Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/nginx_running_on_host georgem' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 386 
Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/nginx_running_on_host vader' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 392 
Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/apache_running_on_host bruno' in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 406 
Warning: exec(): Unable to fork [/usr/bin/ssh synergy /bin/ps ww 10388 ] in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 85 
Warning: exec(): Unable to fork [/bin/ps ww 1709 ] in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 85 
Warning: exec(): Unable to fork [/bin/ps ww 1889 ] in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 85 
Warning: exec(): Unable to fork [/bin/ps ww 2072 ] in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 85 
Warning: exec(): Unable to fork [/usr/bin/ssh georgem /bin/ps ww 21836 ] in /disks/carolyn/b/home/boincadm/projects/sah/html/seti_boinc_html/sah_status.php on line 85

Hmm...

Warning: shell_exec(): Unable to execute '/home/boincadm/projects/sah/bin/apache_running_on_host synergy'

Sounds like that's what was happening to the Seti web server yesterday.

Well, if the have the manpower to investigate and cure the problems, rather than simply apply brute force, there's some smoking evidence here. But I'm not holding my breath.

Errors that show when reporting a post:
Warning: Missing argument 6 for show_post(), called in /mydisks/a/users/boincadm/projects/dev/html/user/forum_report_post.php on line 89 and defined in /mydisks/a/users/boincadm/projects/dev/html/inc/forum.inc on line 493

If 'all your files' includes your BOINC data directory, then make sure the new BOINC install points to the same place, and it'll pick up just where it left off - including work in progress. No attaching needed.

Biggest set of warnings yet [As of 6 Jun 2017, 11:00:03 UTC]: 133 warnings on the SSP. I think I'd better report them in.

It would seem a good thing to do. Thank you.

Done. It appears that during these episodes, the remaining information is unreliable: upload and download servers, for example, are shown as 'disabled' when they are in fact running normally.

Jeff is looking into the SSP warnings, and is asking if anyone saw similar events prior to my report here '[As of 3 Jun 2017, 8:40:05 UTC]'?

I've pointed out that was only 24 hours after a web-server crash and reboot by data centre staff - that'll be what made the messages seem important to me.

They've put in some extra diagnostics and will monitor.

At which point, another email flashes up: "I think we fixed it."