All:
I've been playing with ZM for some time, nearly a year...
Got a PV149 or 150 thingie from BC and a hard wired cam from RS (doesn't matter, it all detects fine, should work)
While I know this is bleeding edged stuff, there is a purpose - security and it needs to work, in a word, critical mass, or folks won't adapt or pay.
I also know, as a sr linux sysadmin, nothing is easy. Most of the leading edge technology is a moving target, ZM is no exception.
The point is simply, ZM is trying to be something for everyone. Considering the obvious limitations of manpower, it' REALLY hard, taking into consideration all the distros. Let the "build from source and here are the dependencies" go (not that there is enough available to comprehend)
I have experienced, first hand, using gentoo, ubuntu, fedora, RHEL the industry is still not ready for prime time in many cases.
What I have learned, is when using the most stable distro, things can go according to (some kind of) plan.
So a little background....
Got etch setup, added a source for ZM in sources.list, installed ZM. A year is like 7 on the internet. Installed ZM v 1.23.X. with relative ease, worked, well, OK.
Time to see if things have improved. Installed Lenny.
The reason?
Debian has proven, time and again, to be a very STABLE distro, unlike ubuntu spinoff of the same - unstable, the issues gentoo has in mgmt, the red-headheaded stepchild RH has divorced - and spun off - Fedora (not to say RH is any kind of distro worth paying for), and yet there are still adjustment in technology that needs to be "massaged" to make life grand. Not for the "average layperson with a spare PC".
And this is sorta the point....
Everyone on the face of this economically depressed world has a leftover, spare pc, that can't run the latest and greatest Microsoft OS and doesn't really care to "shoehorn" it into working cuz it's been a virus magnet that no one can tame, right????
It's all about crime and energy. We want to keep tabs on what's going on. Enter ZM. A pc ready for trash, with a couple hundred dollars and we can feel warm and fuzzy again. Make no mistake, only a few of us are in it for "the thrill of the hunt". The masses just want it to work, at which point, they will pay.
Docs are important. Most don't/can't read (again the masses).
I would pay for a live CD, like Mepis, that is based on a STABLE distro, with a "install me on a HD" button, that just works when it's all said and done. I'm sure many, many would.
Yep, there are lots of variables and it won't work for everyone with a $19 webcam from WalMart, but with some "run of the mill" hardware and $200 or $400 some industry partners working together (like BC and ZM), more folks would jump on-board, write a check for more features.
Just some random thoughts, not meaning to dis on anyone. ZM is a great product, just want to be done in 30 minutes and cover my loved ones, if you know what I mean.
Flame suit on.
Constructive criticism
From what I have seen most issues are:
* ffmpeg problems - every build of ffmpeg is different from the last, they change the header files, and even the options, for instance the latest svn build does not even have the --enable-swscale option.
* automake/autoconf issues - various distros have different versions of autoconf, and the ZM makefile is chock full of autoconf aliases.
* Shared memory - Most people do not have experience tweaking shared memory settings. At least now there is the mmap option, which makes ZM not need to use shared mem, probably at a small performance loss.
I think it would be helpful if we had a group of volunteers to create rpms and deb files for at least the more popular distros. Then people dont have to worry about compiling and just have to satisfy the package dependencies like libjpeg, etc which are easily installed with the distros package manager.
* ffmpeg problems - every build of ffmpeg is different from the last, they change the header files, and even the options, for instance the latest svn build does not even have the --enable-swscale option.
* automake/autoconf issues - various distros have different versions of autoconf, and the ZM makefile is chock full of autoconf aliases.
* Shared memory - Most people do not have experience tweaking shared memory settings. At least now there is the mmap option, which makes ZM not need to use shared mem, probably at a small performance loss.
I think it would be helpful if we had a group of volunteers to create rpms and deb files for at least the more popular distros. Then people dont have to worry about compiling and just have to satisfy the package dependencies like libjpeg, etc which are easily installed with the distros package manager.
-
newvisionantenna
- Posts: 381
- Joined: Sat Jan 17, 2009 7:49 pm
- Location: Germany
Hang tight and I will have an updated version of my Zoneminder install DVD. Working on the last couple of bugs and creating a new user interface for deploying fast wireless mesh cctv security systems.
"The real difficulty I see in this forum is that people are not on the same page.. Everybody is running some off the wall linux distro and trying to build this from scratch, which I can tell is not going very well..
If everybody here were running the same build of zoneminder, installed from the same exact DVD and linux distro, and everybody had a clone machine to the next person here, this thing would take off the grow wings..
But with everybody struggling to just get this installed in the first place, and then if you do get it installed and running, you are on your own because now you have created your own beast and nobody else will know what is going on with your system to help you..
These are my observations while reading this forum over that past several weeks.. This is obviously a kick ass piece of programming, but it doesn't seem to be getting traction because of the reasons I point out above. There is no distribution system, nor is there a standard system to run it on.. Those things are even more important than the software itself, as you can see..
All of the tools are here, and I am thinking that people should just pay a small amount, about the value of one camera, to get this installed and get on the same page and move forward in large numbers..
I have all the tools for that sitting right here ready to go whenever it seems that people will get the drift of that..
Fire away if there's anything I can help you with along the way.. "
"The real difficulty I see in this forum is that people are not on the same page.. Everybody is running some off the wall linux distro and trying to build this from scratch, which I can tell is not going very well..
If everybody here were running the same build of zoneminder, installed from the same exact DVD and linux distro, and everybody had a clone machine to the next person here, this thing would take off the grow wings..
But with everybody struggling to just get this installed in the first place, and then if you do get it installed and running, you are on your own because now you have created your own beast and nobody else will know what is going on with your system to help you..
These are my observations while reading this forum over that past several weeks.. This is obviously a kick ass piece of programming, but it doesn't seem to be getting traction because of the reasons I point out above. There is no distribution system, nor is there a standard system to run it on.. Those things are even more important than the software itself, as you can see..
All of the tools are here, and I am thinking that people should just pay a small amount, about the value of one camera, to get this installed and get on the same page and move forward in large numbers..
I have all the tools for that sitting right here ready to go whenever it seems that people will get the drift of that..
Fire away if there's anything I can help you with along the way.. "
I agree w/both of you on many points.
A moving target is very hard to hit.
I spent more time than I'm willing to admit getting 1.23.3 running on lenny, and it's flakey at best (not so fond of 1.24)
A project director, delegating, asking the right questions, compiling the answers is really what's needed IMHO.
This is about organization, not technology, without red tape, at this point.
Traction, consistency, stability, the rest will come.
A moving target is very hard to hit.
I spent more time than I'm willing to admit getting 1.23.3 running on lenny, and it's flakey at best (not so fond of 1.24)
A project director, delegating, asking the right questions, compiling the answers is really what's needed IMHO.
This is about organization, not technology, without red tape, at this point.
Traction, consistency, stability, the rest will come.