Hi
I have been using the TL Linux app during several years for work on the NSC Triolith and later Tetralith systems. AFAIK NSC never implemented the Web Access part of TL.
Any chance you will release a TL client for ARM64, not HF or consider a release for Android?
1 Like
Hi Peter, as far as I know there is nothing currently on the roadmap, but perhaps one of the dev team can clarify (@samuel)? Web Access was intended to provide ThinLinc access from platforms without a native client available, like Android/iOS. So the path of least resistance may be asking NSC to activate the web access component.
1 Like
@aaron Thank you for replying. I posed the question to NSC about half a year ago and to no surprise they were not interested in exposing another executable to the wonders of internet access - which I understand, completely.
Edit: Perhaps I should re-phrase myself… it is not that NSC were strictly “not interested”… it was more along the lines of there not being strong enough motivators to add the increased risk.
1 Like
Both of these things are stuff we’d like to do, so it’s a matter of prioritization. However, just like @aaron said this is not something on our roadmap currently.
1 Like
It would be very interesting to see the actual roadmap. Do you have anything you can share with the general public?
3 Likes
I am also interested.
Perhaps that was a poor choice of words by me. We don’t actually have a long term roadmap as we want to be flexible to what happens around us.
We do of course talk loosely internally about what we believe we will do within the upcoming year or so. However, since these things are constantly subject to change it’s not something we want to share publicly.
I can however share that we are currently working on the very large project of converting ThinLinc to Python 3. That work can be expected to be finished and released around summer 2021.
1 Like
Seeing this thread here about the TL client and support for other platforms, I want to also express my/our wish to get the TL client to support more other platforms and definitely also aarch64/arm64.
In fact, IMHO it would be a great move - and probably best - if cendio could consider to open-source the client and move development of it to, e.g. GitHub so that interested third-party developers (like me) could contribute / develop own client versions and also potentially merge these changes back via pull requests.
IMHO the ThinLinc infrastructure could benefit a lot from open up more regarding open source and public development on GitHub&Co. And especially the TL client should be somewhat uncritical to open the development, which could help to significantly increase the community around ThinLinc (which is still rather small given the great technology behind this product!).
2 Likes
May I ask back here what the roadmap/priority regarding ARMv8/aarch64/arm64 binaries of the ThinLinc client are? As we are currently moving over from x86-based ThinClients to ARM-based ones (e.g. RaspberryPi4) and would like to use our own thinclient OS (thinRoot) with it, it is curial that the ThinLinc client is also provided as pure 64bit ARMv8/aarch64/arm64 binaries. So please consider investing some time into this as you would anyway have to go the ARM road due to the switch over of macOS from x86 to Apple Mx ARM processors.
And in addition, please keep in mind that it would still be great if you could open source the ThinLinc client at least (e.g. in GitHub) so that third-parties (like me) can compile the client for their own needs at least.
We have a release coming up this summer, after which planning will commence for future versions. Different client architectures will likely be a part of this discussion, but as of now there is nothing on the roadmap. Appreciate your input, though, and it will be taken into consideration.
Regarding a fully open-source client, it’s a trickier proposition since there are other factors to take into account besides the technical ones. I don’t think anyone here is fundamentally opposed to the idea, however it would require some thought (can we support clients people build themselves, for example).
I can fully understand that @aaron and thanks for your prompt reply. And please don’t get me wrong. Its not like that we would start to develop our own sole client for thinlinc in any way. But having the development of the client, e.g. on GitHub would help us a lot in 1. communicating with you in more detail on a technical level as the whole GitHub infrastructure would allow to send in patches as well as direct pull requests. And 2. it would allow us to compile an own version in case you guys didn’t had the time/resources to compile the client e.g. for this platform yet. And 3. I could then also try to integrate compilation of the client directly in my thinRoot project which currently is only able to download your binaries, thus not ARM64/aarch64 support yet because you don’t have a published 64bit ARM version of the thinlinc client out yet.
So my suggestion would be, that you should really consider moving the client development over to GitHub under a strong open source license (e.g. Apache 2.0) to allow third-parties like me to contribute at least a bit to the development of the client. Whether and how you then provide support for these third-party clients (I would not!) or integrate pull requests and changes from third-parties is of course up to you. But it would allow at least to fork your work and suggest direct code changes back, e.g. via pull requests, like this is nowadays done. And to be honest: It still quite puzzels me that there is still no “cendio” organization on GitHub (i.e. you seem not to be quite active there?!?) and searching for ThinLinc on GitHub still provides so less projects with reference to ThinLinc or Cendio which IMHO is very sad given the nice product you have developed all the years. So IMHO you should really consider to open source more components of ThinLinc, starting with the client!
1 Like
There is a Cendio organization on GitHub, although admittedly it doesn’t get much love. Our developers are very active on GitHub under their individual accounts, being maintainers and key contributors of open-source projects like noVNC and TigerVNC. You’re right though, we could do more to highlight Cendio’s involvement in funding this development.