Thanks for the response Henrik and appreciate the confidence with the upgrade license being free.
As for features…
- TLS - The implementation today leaves much to be desired. Configuration of it is not within the UI and is relatively lackluster in the ability to control the cipher sets that are supported by the internal web server that is hosting the services. For example I do not want to be able to support any SSL versions or TLS v1.0 - but would like to provide connections greater than TLS 1.1 (this is pretty standard today given things like the Mozilla TLS profile recommendations: https://wiki.mozilla.org/Security/Server_Side_TLS). Also things that could make implementation of good practice security would be to support initiatives such as LetsEncrypt instead of recommending to customers to buy an overpriced certificate from an organization such as VeriSign. Today a reverse proxy is the only way to really make this suitable for public exposure unfortunately. (disclosure: I am a security engineer)
- CV - The industry is moving towards CV in many respects and simple object detection are becoming more common in competitive products. The ability to recognize people vs pets, cars, etc are becoming the norm. I know there is LP recognition in the software today, although, frustratingly it does not support US plates (probably because of the wide variations per state - but still). The ability to create recording triggers based on some high level CV aspects would be something I would be very interested in if those pieces are in the pipeline. If not within the product I’m wondering if you’ve considered an external pipeline to provide, for example, a video output that has an associated input and API response listener to run the feed or event through custom OpenCV pipeline and back into the platform via the API listener to provide additional metadata to the feed. So… I send a feed to my OpenCV code, that code hooks into the API and sends metadata about the recording to indicate if there was a brown dog in the event. So ultimately expose a way to use Netcam as the central recording authority but allow externalities to decorate the video with more insight in a flexible manner.
- Quick Scrub - One of the features I really enjoy about a competing product is the ability to review all motion events for an entire cam/day quickly. It has a mode where I can just hit play and it will scrub through, at whatever speed factor is set, and I can validate in the course of a few minutes if there was anything I need to go back and review. It’s kind of there with timelapse, but even then the timelapse implementation is frustrating with 1.x because there’s no good way to view it without manually creating one.
- Cloud Integration - While there are some integrations today (Dropbox/GDrive) I’d like a more robust endpoint in the cloud that I can build workflows around. For example with S3 in AWS I can provision lifecycle management and automatically migrate video archives to Glacier for cost savings. Dropbox and GDrive are comparatively expensive and lackluster in comparison of extending my more customized workflow.
- Additional Libraries / Expanded API - It would be great to get a library that is tied to a more flexible language such as Python. In my limited testing I can definitely use what’s there, so I realize this is slightly benign/non-issue, but an official library is always awesome. If anything maybe I end up just writing it and maintaining it on behalf, but obviously a stable API is important. Curious how much this may be expanded / change going forward. And, don’t get me wrong, there’s already a lot there which is very much appreciated.
Finally… What is the ETA of 2.x if that’s kosher to ask publicly?
Thanks so much! And, by no means is the above in any way a slight. NCS seems to be a great product - but a few areas of concern for me before I pull the trigger on buying a few licenses.
Edit: Also - why was this thread unlisted @Henrik ?