No surprise, thereâs a flood of features requests for NextcloudâŠ
AMHO, Nextcloud should stay, in a first stage, with his basic functions and NOT develop a bloated âgaz factoryâ. âBells and whistlesâ could be added way after, but for now developpers should focus only on stability, security, response time (still a bit slower than similar servers), and increased performance.
Two apps only should be improved and get better attention. Calendar & Contacts.
I hope it will not become a âchristmas treeâ or some sort of âswiss knifeâ, doing everything but badly.
However, automatic service discovery support for clients and servers could also be added without interfering very much with the core code, because this should be completely separate from it anywaysâŠ
My feeling is that Nextcloud tries not to produce more features, but changing the way people can interact and contribute with it. See the example Frank was giving in the interview recently comparing Microsoft and Linux. Whereas Linux is growing with its openness.
I think it is quite more important to help other contributing than doing it yourself behind the curtain (to make profit). But this is a question of expectationsâŠAnd we should not forget that Nextcloud just started, development capacity wonât be overwhelming yet.
I think itâs a good thing to see so many users joining up to discuss the features theyâd like to see. They canât all happen, of course, but already this community looks like a far healthier and more lively place than the OC forums.
Yes, Iâd be concerned if developers seemed to be trying to run around developing all of the ideas on these forums, but theyâre not. As far as I can see, users are giving feedback for what theyâd like (yes, in bucketloads) and developers are being careful to balance the needs of the users with the design vision and development capacity⊠just as it should be.
Personally, OC 9 had already almost all the features I could think of needing, beside one: what I requested, which is Single-Sign On support (and someone else proposed in general a pluggable authentication module). I wouldnât want a huge, heavy application (OC/NC are already written in PHP ), but in 2016 having to rely only on LDAP for authentication is crazy.
Ok, you might be right. Iâve proposed some features of the category âbells and whistlesâ also. IMHO I would think of a procedure where a public voting shows what features users would like to see in the future and within what time.
This should not block or enchain the developers but would be an indicator in which direction to go for a good acceptance. On the the users side users would be able to see how âoffâ they are with their proposals.
Adding to @JuergenGrete: Partners could join in to create the stuff that is wanted but not doable by the internal developers in the short term⊠(or maybe creating the ground work (stuff like the horde lib for mail) that can be used in other projects too, and then it becomes a much simpler/faster job to get it into NC).
