Hi guys, I am happy to announce that we are moving slowly, but yet so speedy, that we can present you with ..
Enjoy, and thanks for the patience and understanding :)
Nothing is lost. I understand. But LMRv3 is going to be on the holding track - on a dead end.
I know we LMR are tight on budget. But as you know I call our data a first class citizen therefore the budget shall also be used to migrate all data. We all have to respect if you go ahead and put LMRv3 on a dead end. But I think we should make it better.
Here some thoughts about the migration:
To migrate you want... a) the content from LMRv3 in the format of LMRv4 and ...b) keep links to LMRv3 content unbroken and redirect it to the LMRv4 content.
Shure you are aware that the migration success from LMRv3 to LMRv4 depends on the congruency of the datastructures from both versions. As I understand there is no congruency (this is the scenario below on the right side). But I assume the congruency might be not that bad (scenario below left and center). Shure there is a way to map the LMRv3 datastructure to the LMRv4 datastructure.
Frits, if you give us both datastructures (LMRv3 and LMRv4) Mogul and I could write the migration code that is needed to migrate as much as possible.
Regarding the unbroken links we know the location in LMRv3 and we know the location in LMRv4. Therefore we can setup some redirects based on this information.
In the end you would have all content in LMRv4 and can shutdown LMRv3 using i.e. a Blue Green Deployment strategy.
Please think about it. Our content is value. Do not put it to a dead end.
NilsB, goddammit :D
For almost 2 years I have been working on getting LMRv4 on air. There's no problem in the knowledge, and bright ideas.
There's a challenge in getting things done. If you want to start coding, you are more than welcome to join Hugo, if you (and he) thinks it can make things happen better/faster. I had to cut down on all I wanted to do. That does not mean that I in any way am unhappy with what v4 is planned to be. But I can only expect things to happen if I work this way.
I know it's cheap, but I have to remind you of the situation in Berlin, where we where about to FINISH a robot to a DEADLINE, and you where doing some complex coding for a simple robot that just needed code that could be written in 10-15 minutes in PICAXE :)
This is kind of the same, though I know it's a cheap one to draw. I try to be finished. And I aim for good. V3 is not that bad, and I designed that as well.
I very much disagree in the terms of "dead end". It's like saying that the forum in v3 is a dead end. You can just go there. It's not on the front page, but there are links to it from the front page.
Let's see where we are going; We might be able to intigrate everything, but let us see. The solution I am heading for is really good, and there are no dead ends - only old LMR with nodes and new LMR with other stuff. And in the new stuff, old stuff can be intigrated. And searched.
It is the v4 in the blueprint now (and with feature freeze) or for ever lost on v3 and bright ideas. Wait and see before you dive in - you do not know what it will be like :)
Great to read that the old stuff can be integrated and edited. No dead end then. Looking forward to what comes next. Support where I can. EOF
Ah! Just mention Berlin.. Oh! :D
Sorry - Thanks for offering help, I have let Hugo know!
Cheers my man :)
Sounds good and I am looking forward to V4.
As mogul already pointed out, it would be good to get the user names reserved for everyone, so nobody else can take it.
i like it :) so exciting!
Looking forward to see it going :)
Please make us a mechanism that ensures what we all get the same user names as we have today. It would become a complete mess if we start registering fellow LMR'ians names.
I would be perfectly fine with a random 12-digit user number, but would be a little annoyed to see another guy impersonating mogul.
Hmm, this could be a quick route to street credit, I hurry up and steal some of my heroes nicks... yeah!
Thanks for considering this. (This also goes to lumi)
What I type here can change. But for now the plan is to just start fresh, and let everyone have a go.
In the quite unlike event that someone signs up as another user (using their name) it will be a lot less work for us to settle this manually, than making a system for it. If some should take your name, you write to me, and I'll let the other user know that something is wrong here.
Again, we MIGHT be able to do some checking, but priority is to get v4 up and running - and even if 20 names where hijacked (I doubt that even 1 will be so) - it's still less work for me to settle it manually, than it will be to make a system checking. You know; You might want a new e-mail address associated with v4, and then a system would tell you that you cannot do that (it thinks it's detecting a hijack), and then I'd have to go in and override manually anyway.
Regarding the user numers; I think it's going to be nice to clean up the database, and only have active members in v4 - call it a mancount. And quite an event to have everyone get a new number. We will just have to take it as an experience. And we will know that everyone below a certain number are "from the start of v4".
Again; We might be able to do different/better, but this is lowest common dominator in the rather time consuming task, that we really just need to get done.
I hope you understand :)
Nice!! Hope I can see it complete after I move back to Taiwan.