Should I build a cloud-based dive logbook?

[epicrecluse]

There was an airline website breach a few years ago that siphoned off thousands of payment card details. Root cause was a JavaScript library component compromised by hackers. This was imported into the website and due to the lack of JS security, dutifully sent the details to the hackers. The airline was fined millions. Have often wondered the fate of the incompetent script kiddie who pulled in those libraries.

Our current challenge is ransomware and the usual exploits to take control of the machine. This is why browsers won't be trusted.

Security | Ease of use | Low cost
Pick any two.

Oof, that ****'s sketch. I don't see browsers going anywhere though, people just gotta be smarter about how they do things. External deps should always be vetted, or avoided, whether web libraries or native ones. Lots of other exploits are possible also, I'm sure. Ultimately it's up to the developers to try and build secure apps.

 

[Wibble]

Problem is simply how can you possibly vet an external library and especially an open source library…. Developers are being brought up on the wonders of GitHub and the like.

Examples; a few years ago JavaScript was a PITA to code (still is!) and someone developed the really useful jQuery library. Was a great help for coders. This begat a cottage industry of addons which could do all sorts of things. Unfortunately one unintended consequence was that coders didn’t bother to learn the hard JavaScript to code properly, they simply bloated their site with tons of addons which were frequently unreliable and way more trouble than their worth.

Accessibility went out of the window too as people without the slightest clue coded. Simplicity replaced with bloated libraries again.

It’s a sad world.

Just as well scuba diving’s not like that. Backfinning, pah, I’ll just push myself off the reef.

 

[BlueTrin]

I use subsurface.

I was wondering if it would be doable to make a generic log book template based where all fields and presentation can be customised to store any type of data, then … I got bored and moved on (as usual)

 

[Hoag]

I use an app called Dive Log. I enter the info on my iPad and then, when I connect to WiFi, it will sync with my iPhone. That is about as deep into the cloud as I get for my log.

 

[Damselfish]

I keep my log in a spreadsheet. I keep the spreadsheet in Dropbox. So I can use it locally and it will sync later if I don't have internet. I get all the handy features of the spreadsheet and Dropbox. I can put in as little or much info as I want.

In theory.

In reality, my log is at this point usually just site and date unless there was really something of note. I have a pile of logs printed by dive ops (liveaboards often do that for you) or scribbled on scraps of paper, that haven't made it to the spreadsheet beyond a trip/dates/#dives summary. Someday maybe they will. Or not. I know overall number of dives, what country I went to when, where I stayed, maybe a note on rooms to avoid. Stuff actually useful to me nowadays. Not the stuff they tell you to log in class, and surely not signatures, which most people don't care about after they've been diving awhile.

 

[dmaziuk]

...Though it is true that we can support offline mode for web apps... the issue with it is that it can't be made free for everyone as a native app could, since it'd cost to keep the servers up.

I might try contributing to Subsurface, we'll see. Right now I'm more interested in the write-once-run-anyware capability of web apps, which are becoming more powerful with modern APIs.

You can do local files in javascript: the user would have to "upload" at the start and "download" on every save... ugh. PITA but doable. I would recommend looking at subsurface's native XML for starters -- keeping in mind that any example you'll find will not have all possible fields and the closest thing they had to DTD last I looked was spread over several C #include files... But it should give you some idea of what you're getting yourself into. Take a look at subsurface anyway, even if just to see how much stuff is available in the interface.

Personally I think the problem is dive computer import. I doubt you can do it in the browser, there is no such thing as standard protocol or data format etc., or even kind of data logged (air integration, IMU readings, GPS when on the surface...) Without that, you lost me: the only reason I have a log is because I can download it from my dive computer. I couldn't be arsed if I had to write it down by hand and I wonder how many other lazy people like that are in your target demographic.

 

[NDonahue]

Hi all, super new diver here, so I may be missing something obvious. But I've been looking into the options available in the digital/cloud dive log space, and haven't been very impressed. So considering building an app myself. Wanted to get the community's opinion before getting serious about it. My unfiltered thoughts are below.

Entirely Web Based
The app would run in your browser, so install won't be necessary, and it'd work on any device with a reasonably modern browser, including Linux systems; which tend to be less well supported by such things. It'd progressively take advantage of available browser features to provide the nicest experience possible, without compromising the portability inherent to web apps.

Just A Logbook
It'd just be a logbook, nothing more. Existing cloud based 'logbooks' seem to include full fledged social media platforms, which also seems totally dead in the case of Diviac. More 'features' means more overhead, which means slower progress/fixes/support. Social features will be limited to the ability to share your profile or logs on various social media sites.

Dive Computer Imports
The app would make use of experimental features available in many modern browsers to import logs directly from a dive computer via USB or Bluetooth.

File Imports/Exports
Would support file imports from/to other logbook apps, and from/to CSV for analysis by other software.

Sign Off
Would allow buddies, instructors, etc. to sign off on a log either through a one-use link or in person (by lending them your device).

Feedback
Not sure if anyone would actually be interested in this, which is why I'm posting here. Any feedback is welcome. I'd rather not waste my time building this if it's a dumb idea, so criticism is especially welcome.

Depending on what y'all think, I'd build this as a side project for myself and select interested individuals (free to use). If the app ever received enough interest to be worth opening it up to more people, we'd need to figure out a fair way to monetize it; since a public app would require more resources than I could fund out of pocket.

Have you checked out Subsurface? Not browser based for data entry or advanced features, but basic log is always online. Linus Torvolds was one of the original developers, so pretty strongly rooted in both linux and open source.