I like this update format. We should do one of these every month!
I prefer them weekly TBH
I haven’t been following updates in code for a long time and I was surprised to see that file container version isn’t a number anymore but a kind of hash instead. What’s the reason for this?
Note that safe cat commands in this document still output numbered versions, which is not consistent with the new version scheme.
Apologies for the inconsitencies in the documentation. A full update is a work in progress.
Possibly someone like @bochaco would be best placed to explain the new versioning system. I’ll include an explanation in the documentation at some point.
As versions (tips of tree) can basically fork then versions make less sense. If we go from
etc. then in a concurrent system, it won’t work as 2b does not necessarily know 2a happened or is happening. So using hashes takes away the requirement to synchronise version numbers and allows us to have concurrent updates.
Thanks David. What I was unsure about is what the hash is based on. Is it the content?
Correct, hash of the content.
I can feel the Maidsafe’s fire! I can feel it!
When we can see farming on github?
Thx 4 the update Maidsafe devs.
Keep hacking super ants
I concur with you as a few lines above in the op we have:
(emphasis is mine)
When you see sites with example “maps.google .com”, it is terrible.
Let me explain why:
The most important Info the address bar gives is, which site am I connected too, “google .com” or “goggle .com”? That is why I think that the top name always should come first, example “google.com / maps”. To place sub name first gives logical inconsistency. As if the file Explorer would show documents.c: instead of c:/documents. And if you want to add additional sub names, it just adds to the confusion.
I was just emphasizing the inconsistency in the OP between:
I didn’t want to restart the debate about element order which has been settled (and implemented) by Maidsafe as top name coming last in safe network NRS (and which is the same order as in internet DNS).
Just out of curiosity, your “E2E tests w/ full network” in GH, does the full mean nodes full of data or certain size of network, or something else?
Sorry, that should really be name “w/ split network”. So that’s w/ 33 nodes atm.
I have been lurking for many months after hearing about the project on a Mozilla podcast a couple of years ago. As a potential end-user (and possible farmer), most of the technical discussion is beyond my ken. Nonetheless, this latest update sounds encouraging and I continue to be amazed at the dedication and persistence of the team on what, to a dilettante such as myself, sometimes seems to be a quixotic quest. But then, I’m getting old and crotchety and may not live to see it emerge into the light of day… Keep up the good fight and I will continue to follow with interest.
No it doesn’t. It’s just a matter of where the emphasis is placed. Its analogous to the unix filesystem hierarchy vs. the unholy filesystem mess that is windows. You must be a windows enthusiast.
98% of worlds people are wrong?
Great that we are building a network for the 2% using Linux, it will sure help adoption?
That is not a fair debate, that sentence can be put everywhere and lacks any value, often used to hide lack of good arguments. Sentences like that are very common in todays media and political debate, manipulative argumentation that is just sad for humankind and leads nowhere.
Maybe if they had a good name system, more % people would use it.