Ideas wanted: should we improve the standalone demo.html — and how do you want to interact with the data?

[ch-bas]

The repo ships a standalone demo.html — a single self-contained file (just demo.html + cameras.json, no build step) that lets you search, filter by brand/type/power, sort, and compare cameras offline. It's the zero-dependency way to browse all 2,618 cameras without the hosted site.

The dataset has grown a lot recently — most notably video.streams[] (per-stream resolution/fps/codec) is now filled for ~1,470 cameras, plus normalized codecs, IK/impact ratings, Frigate configs, low-light lux, and more. The demo doesn't surface most of that yet.

So, two questions for anyone using the data:

1. Is demo.html worth improving? If yes, what's missing? Some candidates:

• Show the new stream table (main/sub res·fps·codec) + copyable RTSP/Frigate config on the detail view
• More filters — resolution, codec (H.265/H.264+), night-vision type, vandal/IK rating, ONVIF/RTSP
• Side-by-side compare improvements, CSV/JSON export of a filtered view, shareable filter URLs
• A "recently added" view (the dataset already tracks added dates + an RSS feed)

2. Or would you rather interact with the data a different way? e.g.

• The hosted site at cctv-database.com (searchable/filterable) is the richer UI — is the offline demo still worth maintaining alongside it?
• A small JSON/HTTP API or query helper over cameras.json
• Direct use of data/cameras.csv / cameras.json in your own tooling (Home Assistant, Frigate, spreadsheets)?

If you're already consuming this dataset somewhere, I'd love to hear what you're building and what shape of data/UI would help. Concrete "I wish I could do X" notes are the most useful. 🙏

[ch-bas]

@fvdpol your opinion will be very helpful!

[fvdpol]

Interesting topic; touching a couple of different questions.

In my view the current 'demo.html' web page can be serving multiple purposes:

• a pure demo of how a web page could visualize and present the dataset -- limited value, and does not need to be feature rich; no need for export, compare etc. functions: just show a (filterable) list of data, and a formatted table with the details (almost like a data sheet 😉).
• a different angle would be a local tool to support the data maintenance: create a table/grid like presentation of a number of selected cameras, possible highly differences

If the web page would evolve towards that second use-case, it would not be a demo anymore, but a tool which could be helpful for contributors to spot inconsistencies and help build the json files. -A key aspect is that a such local tool would give a good preview of the data before it gets published on the https://www.cctv-database.com/, allowing for rapid local iterations. So yes, there is value in a (more simplistic) local browser.

Which brings me to your 2nd question, on the data interaction. Before diving into the technical/tooling aspect, there is the fundamental question on why the cctv-camera-database exists, and who will be using it for what.

Focusing on the use-cases and personas involved is key. This will help identify what data is required so serve a certain use-case.

I'll try to give a couple of scenarios that spring to mind, but obviously given more thought that list can be quite extensive.

• When searching the specifications of a certain camera (one's owns or is intending to purchase), the go-to place would be the manufacturer's website and data sheets, rather than any other compiled resource. The sources field in the cameras schema should give this information.
• When researching for a new camera to purchase, a comparison of specifications is very helpful. This will start with a short-list based on what is available in your region/area within your budget constraints: so one will start from vendor/suppliers (who sometimes offer their own comparison/selection tool), and in many cases end up with a comparison table in some spreadsheet covering aspects like price, supplier reputation and a selection of critical specifications for the cameras.
• When adding a camera to your NVR, this will require some kind of configuration. Where commercial NVRs come with a build-in database of supported cameras and corresponding configuration, the configuration in e.g. Frigate can be quite overwhelming and finicky.

Depending on the use-case in mind, a different type of data access (and to different elements) would be required.

In case of the user who want to configure his/her new camera in Frigate, that user would probably get the best experience if from within the Frigate application there would be just a add/configure camera dialog, which can

• auto-detect from the camera's ip-address (or even better: use some discovery protocol!),
• or select manually from a list of vendor/models,
• show information on what would need to be done on the camera local configuration (via link to wiki?)
• provide some camera-specific parameters with sensible defaults.

Can this be done? theoretically yes, but in reality, the biggest blocker here is the lack of know-good configurations: a Chicken 🐔 and Egg🥚 type of problem 😄.

Before wrapping up this long brain fart...
How do I access the data? I use python with the native json library to query the data and extract what I need (often with regex for pattern matching)

[ch-bas]

Thank you for taking the time to share such detailed feedback!

My idea when I started this was just finding a place where I could compare cameras before buying one — and then when I posted it on Reddit, someone suggested adding Frigate support, which turned out to be the best feedback I ever got. 🙌

Your "local QA view that highlights inconsistencies before publish" idea especially stuck with me — that's the one thing the hosted site can't be, and it fits the rapid-local-iteration workflow perfectly. The point about known-good Frigate configs being a chicken-and-egg problem is also spot on; I think a local tool that makes it easy to preview and correct a config before PR-ing it back is exactly how that loop gets unstuck.

You've given me plenty to work with here. 🙏 I'll find some time to put together the next plan for improvements and features, and I'll come back to you soon.

Thanks again for your feedback and your time!

[ch-bas]

Hi @fvdpol , I stared working on your different points and made some progress, I would be happy if you could check the new additions for the offline tool(under tools directory).
Thanks in advance!