Jump to content

Gauging interest for a project idea: a new database website for watch calibers

m0g
 Share

Recommended Posts

m0g Member

m0g

Posted
Posted (edited)

Hello there,

I'd like to gauge interest for a project idea of mine. I'm a web developer by trade and for the moment I'm stuck at home with the good ol' rona, so it got me thinking.

What I'd like to do is fairly simple. Just a new online database for watch calibers. I know how hard it can be to find information on new and old watch movements. And the existing resources albeit great are also showing their age.

So here is what I propose:

I would build an open source web app based on Ruby on Rails and SQL (mariaDB, MySQL, PostgreSQL). Each caliber would have its own page with the following information: brand, model, pictures, year of production, width, height, mainspring reference, lift angle, hand sizes, dial feet position...

Since all the data will be relational, you could browse by brand or by size or width or whatever.

Obviously I cannot gather all this data by myself, I have a bit of experience, but I am an amateur watchmaker at best. I would therefore create formular pages where everyone will be able to upload his/her own caliber information via his/her personal account.

Of course, I would add a basic search engine support.

Currently the online resources I know of:

To bootstrap the database, I would ask those respective website if I'm allowed to scrap their data and of course reference them.

The project would be developed in an open source fashion and published on Github

What do y'all think, would there a need for such or am I just wasting my time.

Is there some things I'm overlooking, functionalities you'd like to have?

Thank you

 

 

 

Edited by m0g
  • Like 3
Link to comment
Share on other sites
watchweasol Distinguished Member

watchweasol

Posted
Posted

Hi  You would never be wasting your time. A single source database in my opinion would be welcomed from that point of view. My own data sources are wide spead and I have to pick off what i want an store in on my system or keep references as to where the data is kept. This only my own opinion there will no doubt be others from our members who may agree or differ but thats what the forum is for the sharing on knowledge and helping fellow horologists.

Link to comment
Share on other sites
m0g Member

m0g

Posted
  • Author
Posted

Hello, thanks for the answers.

The Pocket watch database is definitely an interesting project, I had never heard of it.

For the tech sheets, yes I would add them. In a sense that I would let users upload their own PDFs. Obviously this will need to be done accordingly to copyright laws.

Link to comment
Share on other sites
JohnR725 Distinguished Member

JohnR725

Posted
Posted
On 12/7/2021 at 11:06 AM, m0g said:

The Pocket watch database is definitely an interesting project, I had never heard of it.

For the tech sheets, yes I would add them. In a sense that I would let users upload their own PDFs. Obviously this will need to be done accordingly to copyright laws.

One of the interesting problems with the pocket watch database is the source of the material listings come from the various watch companies parts books. But a lot of times the parts book consolidate watches into a category and that does not tell the entire story of the various watches. Or simplistically the database is not perfect.

Then there's the problem of how much storage space are you going to have for this project? In particular for the tech sheets because one or two tech sheets not too bad but accumulate thousands of them and you're going to have gigabytes of data.

Then there's the other thing is important to have would be cross-referencing. Cross referencing is needed because we end up with basically OEM movements that are used a whole variety of companies watches all with company names they conceivably all cross reference back to one movement. That's why companies such as bestfit existed. They produced printed catalogs and now they have an online presence. Unfortunately you do have to pay for this one.

http://bestfitmaterial.com/

1 hour ago, TimpanogosSlim said:

I have often wished it had parametric

They do have this search which last time I used it I didn't find to be very helpful

http://www.ranfft.de/cgi-bin/bidfun-2uswk.cgi?1&ranfft

 

Link to comment
Share on other sites
JohnR725 Distinguished Member

JohnR725

Posted
Posted
1 hour ago, LittleWatchShop said:

You are so last century.  We don't use that any more.  Terabytes is now the fundamental unit of storage.

I suppose I should blame Microsoft but they would be mad I'm still using Windows 7 on this machine. The reason I made the reference to quantity was that I was looking at one of my folders of horological data and it was. There providing the website did the math correct that I looked at that's why I was commenting will take up a lot of space. For chat like now were talking terabytes I assume the cost is gone down because the old days gigabytes of space would cost money.

 

22 GB = 0.022 TB (in decimal)
22 GB = 0.021484375 TB (in binary)

 

 

Link to comment
Share on other sites
TimpanogosSlim WRT Addict

TimpanogosSlim

Posted
Posted
4 hours ago, JohnR725 said:

One of the interesting problems with the pocket watch database is the source of the material listings come from the various watch companies parts books. But a lot of times the parts book consolidate watches into a category and that does not tell the entire story of the various watches. Or simplistically the database is not perfect.

Then there's the problem of how much storage space are you going to have for this project? In particular for the tech sheets because one or two tech sheets not too bad but accumulate thousands of them and you're going to have gigabytes of data.

Then there's the other thing is important to have would be cross-referencing. Cross referencing is needed because we end up with basically OEM movements that are used a whole variety of companies watches all with company names they conceivably all cross reference back to one movement. That's why companies such as bestfit existed. They produced printed catalogs and now they have an online presence. Unfortunately you do have to pay for this one.

http://bestfitmaterial.com/

They do have this search which last time I used it I didn't find to be very helpful

http://www.ranfft.de/cgi-bin/bidfun-2uswk.cgi?1&ranfft

 

Regarding space, I'm under the impression that it is only getting cheaper. 

But my bias is that one of my siblings is a principal engineer at Amazon, and another one is high up in the enterprise support structure in a company that sells NAS systems. 

I hear that my ISP (XMission Internet) is excellent too, but i have never had an idea worth paying for hosting. 

I had forgot about that search screen - been years. I kinda wish there were a simple way to restrict by size, complications, etc. I dunno, maybe I'm underthinking it. 

  • Like 1
Link to comment
Share on other sites
  • 2 weeks later...
spectre6000 Distinguished Member

spectre6000

Posted
Posted

Former Rails dev here (pretty much exclusively Python on hardware for the last few years). Wife works at a Rails shop as well.

Whiteboarding here:

I like this idea. Organization will be key. There are a number of movement-oriented reference sites out there, and they all fall short in their own ways. Getting it built is its own challenge, but I think success is going to hinge on design with this. A lot of the old school watchmakers with the knowledge you need aren't very technical in a digital sense, and a lot of the new school guys are, but will mostly be consumers.

There are a number of paradigms from which data could be sought and presented. Movement identification is an obvious one. Movement data once you know what you're dealing with is another. Movement interchange is something people often want to know, but data is not very available. A photographic database of individual parts by movement to go with service sheets could also be extremely useful. A marketplace could be a source of funding to at least keep the lights on, maybe a small staff of engineers; parts, movements, cases, etc. 

Brainthinking here?: It may be worth trying to organize the data simultaneously deep and wide based on user preference.... Not sure how you'd overlay that.

You're deep in wiki territory here though. There's no way one person, or even a nominal team, would be able to manage this data. I know that's pretty much a domain unto itself...

Digging the idea. If you want help, let me know. My Rails is rusty, and I've never worked with mariaDB, but I imagine the overlap of watchmakers and (even former) Rails devs isn't huge.

Link to comment
Share on other sites
LittleWatchShop Distinguished Member

LittleWatchShop

Posted
Posted
20 minutes ago, spectre6000 said:

If you want help

I will help. Though I have written lots of code over the years...it was all engineering based and not the kind of stuff we need here.  My son scraped ranfft with python using beautifulsoup and some other packages I think--I have his code.

Funny thing I use the EVERYTHING search engine on my computer.  Now when I type in ETA, I get the whole list of ETAs from the database.  Then I click on either the images or the text information.  Crude but cool.

Regarding Images, everytime one of us services a part, we at least take a high res picture of the movement, ranfft style.  Sure, getting a picture of all the parts is even better, but may be hard to coax everyone to do that.

I am going to start doing the ranfft style pictures going forward.

Link to comment
Share on other sites
TimpanogosSlim WRT Addict

TimpanogosSlim

Posted
Posted (edited)

After i fumbled my L.G. #350 mainspring winder today, resulting in an AS 1701 mainspring with numerous creases in it somehow, I discovered that the mainspring data on dr. ranfft's site doesn't directly translate to, say, otto frei's available mainsprings. 

aside from metric vs. english length, what the heck is a denison? Why don't we measure mainsprings according to the average width of the stem of a specific wildflower?

Are there european and american standards for mainspring dimensions? 

Regardless, ultimately i discovered that Bestfit WA32 fits this movement, according to Jules Borel, who would want $28 for it if they had any. 

Also, i discovered that, like the stem, it shares its mainspring with the AS 1580. I ultimately paid $12.78 shipped on eBay. 

I don't know. Maybe a feature could be the ability to add 3rd party part numbers and parts interchange, and for people who did not add that data to vote whether or not it is accurate? 

Edited by TimpanogosSlim
Link to comment
Share on other sites
JohnR725 Distinguished Member

JohnR725

Posted
Posted
19 minutes ago, TimpanogosSlim said:

aside from metric vs. english length, what the heck is a denison? Why don't we measure mainsprings according to the average width of the stem of a specific wildflower?

Wouldn't it have been better if they had just engraved the mainspring dimensions on the main plate of the watch?

you might find this link of interest

http://www.watkinsr.id.au/mainspring.pdf

Then always a good practice when replacing components in a watch is to measure the old and measure the new and make sure that they appear to be the same to avoid unpleasant surprises. This is actually very common when replacing American pocket watch staffs. Because one number could refer to a whole bunch of different staffs so different that they all should have been given separate numbers but they were not.

34 minutes ago, TimpanogosSlim said:

I discovered that the mainspring data on dr. ranfft's site doesn't directly translate to, say, otto frei's available mainsprings. 

I don't quite understand that? If I go to

https://www.ofrei.com/

Being lazy I just search for 1701

Came down the list until it said mainsprings found the third link.But also added in the first and second link it looks like all the dimensions are pretty darn close to each other so I don't quite understand the confusion here other thanThe link to PDF above that I attached. One of the problems and watch repair and even watchmaking is the lack of standards and the ability to adapt other people's measurements. Which is why mainsprings can be referred to with metric simultaneously with inches and Dennison all on the same package. Watchmakers adapt and use everybody's measurements sometimes all at the same time.

 

http://www.ranfft.de/cgi-bin/bidfun-db.cgi?10&ranfft&0&2uswk&AS_1701

Zf675, 1.20 x 9.5 x 0.10 x 315mm

http://cgi.julesborel.com/cgi-bin/matcgi2?ref=q}B~UVG

MSS-WA32

 MAINSPRING 1.20-.105-340 10

 GR 2937-X

https://www.ofrei.com/page1265.html#19047

Watch Mainspring, Automatic For AS 1700 and others

Metric Width: 1.20 mm, Metric Strength: 0.105 mm, length 340 mm barrel diameter 10

 

 

Link to comment
Share on other sites
m0g Member

m0g

Posted
  • Author
Posted (edited)

Hello, and thank you all for your interest.

Sorry for the technical jargon. If something is unclear, I'll be happy to clarify.

Here are a couple of thoughts in no particular order:

 

Regarding movement white labeling (example: the Gruen 455r is a rebranded AS 1758):

For this I would add a parent id in which a caliber can refer to another.

 

Regarding Movement family (example: the eta 2824 belongs to the eta 28xx family):

For the family reference, I could create a specific Family table for which a movement can be referred to. Which also means that each family would have to be created manually.

 

As for technical spec sheet and individual picture of parts: I'd like to try to keep things as simple as possible for now. What I will do is the following:

* For each caliber, one can upload pictures with a legend underneath.

* For each caliber, one can upload PDFs files

 

Movement interchangeability will be complicated, for me the critical data is:

* Dial feet position

* Stem height

* Stem width/diameter?

* Hand diameter

* Width

* Height

Based on this you could swap movement around between watches (correct me if I'm wrong)

 

Also, at the moment, these are the fields I have in my database for the movement:

* Name

* Width

* Height

* Lift angle

* Manufacturer

* Stem height

* Jewels count

* Start production date

* End production date

* Has date?

* Has day?

* Is automatic?

* Has hacking?

* Is chronograph?

* Hour hand diameter

* Minute hand diameter

* Second hand diameter

* Beat rate

* Dial feet position

 

For the mainspring, I could either create a new mainspring table or add it directly to the movement table. As for the fields, it would use the following structure (taken from https://watchguy.co.uk/cgi-bin/mainsprings):

* Manufacturer

* Caliber(s)

* GR code

* GR reference

* Height

* Thickness

* Length

* Diameter

* Type: manual or auto

* Material

Edited by m0g
Link to comment
Share on other sites
spectre6000 Distinguished Member

spectre6000

Posted
Posted

You have boolean values for a handful of complications, but there are a lot more complications than that. 

Dial feet position will be challenging. How will that be recorded?

Values/specs for replaceable parts; ie. mainsprings (touched on), stems, I feel like I'm forgetting at least one more, but maybe those are the only two big ones.

Might break auto out by type: rotor, bumper, microrotor.

Probably more, but toddler duty...

Link to comment
Share on other sites
  • 1 year later...
m0g Member

m0g

Posted
  • Author
Posted

Hello,

I just wanted to let you all know that I'm not giving up on the project. It's just that it's taking a lot of time and progress is really slow.

Here is a screenshot of the caliber page.

The app will support both mechanical and quartz movements.

Still a lot to do. I'll share a URL with you once I have made more progress.

Cheers

image.thumb.png.66b91285c17d4183262fe8e27842d264.png

image.thumb.png.b8996a31f3a04917dc463d9b75091145.png

  • Like 1
Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
 Share
Go to topic listing


×
×
  • Create New...