Hi, as announced here, the rowing data and analytics site rowsandall.com will disappear a year from now. The idea is that most of the rowing related functionality will be implemented on intervals.icu with help from David and myself, but hopefully also from this community.
I thought it might be a good idea to open this topic so we can discuss.
Looking forward to the integration/migration.
Hi, looking forward to the migration. As a start I’m familiarizing myself with intervals.icu and trying to export my data from rowsandall to intervals.icu. The rowsandall website tells me it is processing in the background but I never see the activity appear here. Is there anyway I can debug what is going on?
Note: I do have activities coming from Garmin in intervals.icu. The data in rowsandall comes from CrewNerd and RP3. So the activities may overlap in time and date. Could that be a problem?
So then is the eventual goal to get intervals to support the RowingData file format? My understanding is that your platform was one of the few that made full use of that data
How can the community help?
OpenRowingMonitor uses the RowingData format as well, as it was very well-defined. In essence, it is as close as native output goes for ORM: other formats are not as rich and thus cut down on a lot of usefull metrics ORM produces.
As we already have a working interface with intervals.icu (currently uploading fit-files), I’m looking forward to switching to the rowingdata csv. For us it would be a single line of code that has to change (and a lot of testing).
PS: I will miss the old RowsAndAll website. It had its issues, but a nice clean interface.
Quick update. David has some deadlines on other projects, so the Rowsandall project is in the queue. In the meantime, I have decided to make the Rowsandall code Open Source on sander/rowsandall - rowsandall - Wereldraadsel Git - Be kind. It started as a “exploring Python/Django project”, then expanded in many directions. Hopefully the Rowsandall source code can be useful for community support on intervals.icu.
One way the community could help is with listing which rowing related measured quantities you feel are important. I have a long list that is supported on Rowsandall.com but we may need to prioritize a bit?
I’ll chime in - for me the top priority would be direct (or as direct as possible) import of data from NK (Speedcoach) to Intervals, like Rowsandall already has. Second priority would be Time measured Courses and Course Based Challenges. Sorry but I am no help with the coding… but I might be able to get you the name of the right person to talk to at NK?
Hi Greg, good to see you over here. 
I know who to contact at NK side, and the NK support is not rocket science. But with that import come fields like catch angle, finish angle etc if we support the Empower Oarlock on top of the SpeedCoach, and we need David to weigh in on that.
If you look at indoor rowing data, I think strokerate, distance per stroke, work per stroke, drive time, recovery time (and thus stroke ratio), drive length, peak force and the force curve are the critical ones.
For OpenRowingMonitor those are the metrics requested most often and that are part of our main GUI.
I had been noodling on this as I was going over workouts throughout the month. Came here to deliver a list. But Jaap covered literally everything I could think of that wasn’t already in Intervals
CrewNerd now exports to Intervals.icu directly.
That’s great… for OTW rows; now we need to get BoatCoach onboard.
@Sander_Roosendaal I’m considering expanding the FIT-file definitions to include the fields mentioned above as developer fields (note: stroke distance and work per stroke are already Garmin fields I believe). Would that simplify your life?
I think that’s more a question for David Tinker @david
and Dan Eiref @ boatcoachapp@gmail.com
We probably need to add built in support for common rowing metrics to Intervals.icu, then we can pickup these from the fit file or compute them if possible. Things that can’t be computed from the records would need to be included in the session. Also need to add activity streams (traces) for rowing specific record fields that can’t be computed.
Need to try make some “rowing time” soon.
Hi
I have rowed on a concept 2 since 2006 and now have a model D with PM5.
I used to row on Rowpro and analysis followed on the rower csv export which I took into xls.
Could the intervals rowing analysis use the PM5 fit file as its data? Nice HR data , stroke count for each interval, stroke force curve, stroke length and av watts are all wonderful.
Really looking forward to the increased functionality - Can you confirm the data will be from linked apps eg EXR and PM5
Regards
Huw
Yes, the FIT file you download from the C2 Log can be uploaded to Intervals.icu. Intervals will report that “This activity has been recorded with approximately 4 seconds per data point. Intervals.icu works best with a data point every second…” but all the functionality still works, maybe with a little less “precision” but it still works.
We should urge C2 to get ErgData to record data at 1s or perhaps get Intervals to re-sample to 1s when is sees a C2 FIT file being uploaded.
Paradoxally, the opposite is true! C2’s approach is actually MORE precise, not less. And it would be stupid to ask them to do it differently.
With the development of OpenRowingMonitor we experimented with time and event based reporting and recording. Our conclusion is that event driven approach is far superior.
The key issue is that rowing is a discontinuous sport, where speed and power change quite considerably throughout the stroke. If you’d see the actual (instantanuous) speed on the monitor or in datafiles, you’d go nuts due to that jojo effect.
So in rowing, all speed/power/etc. data is calculated and reported once per stroke (aside from time and distance travelled) as that is the only relevant basis of comparison. A PM5 thus also reports that data when the strokes completes: it is stroke driven reporting, as rowing is an stroke driven sport and all reports cover that stroke.
When you start reporting on a time driven basis, you actually lose crucial information, like stroke timing and work per stroke, which can be deduced from accumulating metrics in the fit-file. And sites like Strava completely mess up speed/pace metrics and think I exploded at the finish with specific drive timing. So intervals.icu is completely wrong here.