OpenWrt and GitHub vs. Gitlab

[Fernando Frediani]

What I am going to say about GitHub some people may not like, but I 
think it should be stopped to be used for the simple reason they don't 
have yet support to IPv6 and don't even seem concerned to have it 
anytime soon.

Some may think that is just a small or cosmetic detail, but I think a 
way to get IPv6 available to a higher number of content is to simply 
refuse use those who are late to implement it. There are no excuses to 
not have it anymore, more or less complex it may be to have it 
implemented, not in late 2021. Most bigtechs with large and complex 
environments like Google, Facebook, Netflix, Cloudflare, Akamai all have 
full IPv6 support, so there aren't much excuses to not have in any other 
scenarios

Some may just want to get their development work done using the 'best' 
tools and not care about the rest but I personally believe each one who 
believes this is important must give the example and refuse tools which 
neglect IPv6 until they understand the message. Sometimes we need to be 
prepared to compromise on something to get something more important in 
long term. And to be honest in this case is it not a big deal as there 
are other great tools to be used. GitLab for example supports IPv6 already.

Finally to finish a last note: in the case OpenWrt moves away from 
GitHub it should make public the reasons so perhaps the message gets to 
the right people quicker.

Regards
Fernando

On 10/06/2021 15:29, Luiz Angelo Daros de Luca wrote:
> Hi,
>
> OpenWrt already depends on github for packages. OpenWrt success
> depends on packages repo.
> The question is if we should keep using it. The two code base with two
> issue tracking solutions is terrible.
>
> I think that there is no problem in using any provider as long as
> OpenWrt is resilient to an aggressive takeover.
> If all codes are signed, externally verified and there is an escape
> plan, it really does not matter who hosts the project.
> As long as we control DNS and have backups, we can bring everything
> back to work. I'm OK with using github even for
> the main page (it looks like it supports custom domain name).
>
> It's simple math: if the cost to migrate to another host is less than
> the saved resources over time (time and money), it is still fine
> to accept the risk. I really think that the energy used to maintain
> OpenWrt infrastructure could be better used to develop OpenWrt.
> Let them pay the bill. We just need a good and quick recovery plan
> like "migrate from github backups to gitlab" (or anything available).
>
> My 2 cents,
>
> ---
>       Luiz Angelo Daros de Luca
>              luizluca at gmail.com
>
> _______________________________________________
> openwrt-adm mailing list
> openwrt-adm at lists.openwrt.org
> https://lists.openwrt.org/mailman/listinfo/openwrt-adm