Forums/Basis Community /Community Help

Any word on export/api?

Nicole Aptekar
posted this on March 21, 2013 12:25 PM

It's been four months since the posted answer of "we're looking into it". Even a super-basic csv would be a good start!

 

Comments

User photo
Linda Gilpin

Ditto!!!!!

March 22, 2013 06:19 AM
User photo
Edrabbit

+1 waiting for API/export functionality.

March 25, 2013 11:13 PM
User photo
Ken Snyder

I think an API is essential for the enthusiast crowd which is a big portion of your initial audience.

March 29, 2013 08:58 AM
User photo
Erik Dahl

I completely agree. Opening up the data to allow individual users own their own data and export it would only be beneficial to Basis. Allow your user base to own and manipulate their own data and they/we can experiment with it and you can then leverage our experiences. It's a win-win to open the data to the owners of the data. 

March 30, 2013 05:09 AM
User photo
Craig

The data is already there! How about just giving us access to the numbers?

April 01, 2013 08:10 AM
User photo
Laughlin Elkind

With an API the usefulness of Basis could be expanded so much, and at so little cost to Basis Science Inc.! 

April 01, 2013 03:35 PM
User photo
Jonathan

This option would be a make or break for me also!! I would buy this device in a "heart beat" (no pun intended)  if you could provide us with the ability to download our raw data. I know that jawbone UP does this option in CSV format. I would like to be able to make use of the data in a way that would be beneficial to my needs.

April 04, 2013 07:33 AM
User photo
Laurie Frick

You could get more enthusiastic support from your early-adopter twitter community by offering data export. Start with basic CSV download. I posted a pic of my new watch on facebook over the weekend and got a quick negative from a very influential quantified-selfer about no data export. You need to fix this before you see a more rampant backlash. Haven't taken the watch off since it arrived (ok, showering it's off). thanks! From an enthusiast.

April 08, 2013 06:30 AM
User photo
Linda Gilpin

As an "early-adopter" I am generally an enthusiastic fan of the Basis band and its potential. I am a bit frustrated by the inability to wear it during water aerobics classes or while swimming, but my only BIG issue is that I can't access and work with the extensive data the band is capturing. I hope you guys fix this very soon, as I'd rather not have to switch to Jawbone Up, bodymedia, etc.

April 08, 2013 09:58 AM
User photo
Beau Gunderson

+1, would love to get at my data.

April 09, 2013 04:27 PM
User photo
Michael C Walker
If I can't get my data in my app of choice, I like many others will choose a different product
April 13, 2013 12:05 PM
User photo
Damien Blenkinsopp

+1 for this - essential.

April 24, 2013 03:09 AM
User photo
Martin Norland

YES! + >9000 !

I missed out on the window in December, and just got the notification that they'll be available soon, so was looking into it again - I have to agree - I don't need anything fancy, I don't need an API to interface with the mybasis site and cause you guys support headaches, I just want to start tracking my own personal analytics.  I'm thinking along the lines of the data enthusiast, not unlike http://blog.stephenwolfram.com/2012/03/the-personal-analytics-of-my... - but even more 'personal'.

Whether it's a dump from the website, something I can pull off the band directly, bluetooth sync'd to my phone, or something that "just happens" as part of the sync to your website, I would love to have the raw numbers to merge with other data I'm tracking over time.

This was what held me up back in December, and it will hold me up again now - especially as my Pebble should (fingers crossed) be arriving in the next month - but I will wear two watches, onlookers be damned, if I can get the raw data from the watch (as much as it's storing/tracking, no more, no less).

April 30, 2013 12:05 PM
User photo
Kyle Smaagard

YESYESYES...

 

Thinking of switching to Pebble...doesn't offer as much (in terms of sensor data) but now they have Runkeeper integration, w00t!

May 07, 2013 12:00 PM
User photo
Laurie Frick

Hi Jeff H. are you watching the forum boards? You are in danger of losing your early adopter community. You need to COMMUNICATE your data export plan. We will be a patient, loyal group if you let us know how you plan to enable data downloads. Start here. 

May 07, 2013 03:04 PM
User photo
Thomas DaMonkey

I just received an update saying more basis bands would be available for purchase soon. However, without export or an API, I'll pass.

May 07, 2013 10:11 PM
User photo
Laurie Frick

Yes, people ask me all the time about the Basis Watch, cause I wear it around the clock. I'm going to begin telling them to wait and look for other products. How a company captures, shares and handles the privacy of personal data will set them apart. 

May 08, 2013 04:44 AM
User photo
Lee

The recent closing of Zeo's website has reemphasized the importance of this to me. It's not just about wanting freedom to crunch numbers however I please but also because I want to make sure I have the date stored personally so I don't have to depend only on a 3rd party.

May 08, 2013 05:38 PM
User photo
Kristin
BASIS

Apologies for not chiming in sooner. We are listening! Releasing our API to the public is on the roadmap but the timing is TBD. 

Please keep the feedback coming! It's important for us to understand our customers' needs and interests so we can prioritize our product enhancements. 

I'll be sure to pass everyone's thoughts back to the rest of the Basis team. 

May 09, 2013 01:58 PM
User photo
Skyler Thomas
Laurie, I am feeling the same way. I was in Vegas last week at a conference and the Basis questions were coming at me nonstop. The lack of a data API is unacceptable at this point (Lee's point about the Zeo situation is spot on) Kristen - I understand Basis' current position. Only you guys can determine priorities and resource allocations. However, my urgent message to you is that you are offending the early adopters in this space. Open data is a non-negotiable item. One of the primary reasons I went with Basis over the competition was the competitions stance on consumer data. The simple fact is that Basis is out of the box less integrated with data services like weight, calorie intake, etc. than any other device on the market. Worse, we are unable to roll our own. I am very well aware that for Basis to work as a company it must appeal to a mass market audience and not geeks like me. However, I believe you have no chance of exciting the mass market if the QS geeks start throwing you under the bus. The smart watch competition is too hot right now. We are currently saying that its early days. My read on the situation is that you have at most a month or two before that recommendation moves to, "forget about Basis, wait till some of the sensors are integrated into other platforms." Your first export does not have to be perfect, a quick CSV with the data contained on the details screen would be enough. I will caution you that this level of detail will not be enough medium term. We will expect and demand every piece of data your sensors are uploading. (Note: I said medium term. You do not have a long runway there. It is my hope that TBD means you have already started considering a detailed API because you are running out of time) I am sorry if this post seems overly negative but this is a VERY serious issue.
May 10, 2013 12:20 AM
User photo
Nicole Aptekar
Huh. Kristen, I find that answer unacceptable.

There is no use for this device if I can't at minimum export this data. It is the most aggravatingly full-of-compromise device out there. This abysmal response makes me glad that I didn't stop carrying my fitbit on me.
May 10, 2013 05:12 AM
User photo
Laurie Frick

Skyler - totally agree. People ask me all the time about the Basis watch on my wrist. And same here, kept wearing my fitbit. Not sure this clunky thing is worth it, the wrist if valuable real-estate, and without a timeframe for knowing when I'll get data download, I'll take it off. My hope is the early adopters can create enough of a stink to get it on the roadmap. We got the response here on the forum cause I started tweeting @mybasis. And said they'll lose their early adopter community - but this is an incredibly weak corporate-ish response, coming from a tiny company. Geez.

May 10, 2013 05:31 AM
User photo
focalintent

Another voice for the "want data access".  I understand the desire to craft a well formed API, believe me, as a library maintainer I certainly do.  

Until then, though - just provide a way to get a dump of data.  That's a simple api - provide a start time, and an end time, and then the raw data.  There's a class of people out there who will happily work with and build things on that raw data - things that you may not even be considering as options while building the API.  Let us at it!

I'm seriously going back and forth on whether or not to take advantage of ordering a basis when they become available again - primarily because I want to be able to pull down raw data, and I'm disappointed to see that it is still not available.

May 10, 2013 05:01 PM
User photo
Leonard Lin

I probably should have posted this a while back as I discovered this months ago.  While this doesn't help non-developers, but for those that are, you can trivially extract JSON formatted data from the API. Just open up the Developer Console in your browser and watch the /api/v1 XHR calls to extract any of the data you want. It's quite nicely formatted.  Someone could pretty easily write a wrapper tool for less technical people if they wanted to as well...

May 12, 2013 05:06 PM
User photo
Lee

Thank you Leonard for this lead. I was able to access some of my data (I think I did it slightly differently, didn't fully understand what you described). Only was able to access one day at a time but here's what I did

In chrome go to the details view fo your data

In Chrome Click View> Developer> Developer Tools

Click Network

Locate api/v1/chart

Open that in a new tab and then you have text you can copy into another file to save. Here's what my data from today look like http://pastebin.com/ntveLYdG

Going to be a bit of work to go through and download this for each day but it's better than nothing.

May 12, 2013 05:46 PM
User photo
Leonard Lin

Yeah, like I mentioned, trivial. Be careful not to post your actual chart URL btw, as the unique hash is the extent of access control for your data.

May 12, 2013 05:57 PM
User photo
Lee

A note to anyone who is going to download data using this. It is defaulting to starting each day at 9pm the day before and ending at 2:59 am the next day. I think some of the attributes in the URL need some fiddling with.

 

Probably something in this part: units=ms&start_date=2013-05-12&start_offset=-10800&end_offset=10800&

 

Don't have time to play with this right now, but just letting everyone know that if you download each day individually right now you'll get some duplicate data due to this overlap

May 12, 2013 06:52 PM
User photo
Lee

Never mind :P I figured it out. If you delete &start_offset=-10800&end_offset=10800

From the url you'll get the correct number of minutes in a day in your output. 

May 12, 2013 06:56 PM
User photo
Beau Gunderson

I also played around with getting raw data from the device using a man in the middle attack. The Basis Sync software verifies the SSL certificate, though, so that was kind of a non-starter.

If you look at the Basis Sync log files you can see more of how the API is laid out--you can get access to goals without an access token, for example.

One possible way forward is to create our own data access driver like libfitbit: https://github.com/openyou/libfitbit

May 12, 2013 07:01 PM
User photo
Damien Blenkinsopp

Hey guys,

Quantified Bob wrote a script to access the JSON data and download it. You can see the details here: http://www.quantifiedbob.com/2013/04/liberating-your-data-from-the-...

The final script you can use on your server is here: https://github.com/btroia/basis-data-export

May 12, 2013 08:53 PM
User photo
Martin Norland

While it's good that one can at least gain access to their data in a usable format - this is still, to my mind, non-preferred for one gigantic reason - simply, it's pulling the data not from the watch, but from the mybasis servers.  As Lee alluded to with the shutdown of Zeo - this watch has a closed upload process to external servers, which then have an export/api - that still leaves any investment in the technology from a long term standpoint a question for the end user, since the data doesn't reside somewhere entirely under their control.

The basis connects to your PC and uploads the data securely to their servers, behind the scenes.  The proposed/requested data dump/export process (again, does NOT need to be an API, just raw dumps in some usable format, say stored in a directory in the mybasis software folder) needs to happen locally on the end users PC.  Even the option to POST to a different server would, at least, leave one with a way to get their data from their watch 10 years from now when mybasis no longer supports this device, because they've suspended support for all non-implant biometric devices.  A wild scenario, but certainly a more preferred option than the route Zeo went.

As long as the format of that local data is relatively consistent (which, seeing as the monitors don't change, seems very likely) I don't think anything would need to be done other than for it to exist before the upload-to-server step, though I would imagine preferably written as a separate file for each transfer (or day, whichever is easiest).

This is by no means a prediction or fearmongering of Basis shutting down, but in all likelihood something will eventually change that causes to the data from this watch to no longer be available from mybasis's servers, and I always try to ensure devices I acquire are not forever dependent on a third party to function (unavoidable for some, such as a cellphone, but I can still use my old iphone w/ the cellular disabled as an ipod touch, just as an example).

May 13, 2013 07:18 AM
User photo
Kristin
BASIS

All,

We appreciate your passion around this topic! We take our customer feedback to heart and understand that data portability is important. We have other popular requests including Android and iOS mobile apps, expanding habits, building additional metrics and providing more strap options. 

We are working hard to deliver new features as fast as we can. However, we are a start-up with limited resources, so it's important for us to evaluate which features make sense to deliver first. Based on the feedback we've seen, we are prioritizing the mobile app and dashboard improvements over an API or data export capability.

Basis is committed to helping our customers with data portability, but we're unable to provide timing until we have more confidence in our delivery date. We have communicated delivery dates prematurely in the past and have learned our lesson. I hope this helps to clarify our reasoning.
May 13, 2013 04:58 PM
User photo
Eugene Bogorad
May 15, 2013 03:20 PM
User photo
Lee

Hi all,

My brother was able to write a quick script using the api to pull multiple days of days of data at the same time and organize the minute by minute data for each sensor into a csv

 https://github.com/dunn5/MetricsCSV

I'll try to write up a readme later on the github with better directions  about using it, but the short version is that you take the unique identifier from the API url and put it in a hexkey.txt file and then run the python file. Enter how many days back you want to go and it will pull those days into a csv file. Each day will be one row. It does not pul the descriptive stats or the data with the unix time codes telling when types of activity intensity occurred mostly because it took some convincing to get my brother to help and it didn't make sense to ask him to pull the parts not relevant to my particular interests. If it were something someone else cared about though it shouldn't be hard to modify this script to include those.

It's a quick and dirty approach but accomplishes a large percentage of what I wanted in data export. Hope it helps someone.

May 15, 2013 05:15 PM
User photo
Letmein

Kristin I sympathize with you and still love what the Basis watch does right out of the box.

 

I also have to say that I knew from the time I purchased that some of the export options would be well down the road, and I'm willing to wait.

 

Even though for me, what seems like questionable timetable choices, doing an Android App before an iPhone App surprised me, I knew these things were all going to take some time.

 

This is a great device, with or without data export for the moment. Providing as realistic as possible timelines as to when some of these features are likely going to be available, will make everyone feel a little better likely for making the initial investment.

May 26, 2013 12:04 AM
User photo
Kristin
BASIS

@Letmein

Thanks for the understanding note. Our teams will have a better sense of the API/export schedule after we ship the Android and iPhone apps. My best guess would be end of year, but we've learned not to promise features until we've nailed down the schedule. 

May 27, 2013 02:01 AM
User photo
Alex P

I got an email invite today and was really excited to come purchase the Basis.  In fact, I was at checkout ready to type in my info and decided, just to be safe, I'd look at some of what the community was saying about it expecting nothing but good things.  However, after seeing some of the crap this company has put out I'm hesitant to order.  Your app is only available on a few android devices, Basis can't accurately measure your heartbeat while you're exercising (http://support.mybasis.com/entries/23438713-Measuring-HR-during-periods-of-even-modest-arm-movement) and you can't even simply get your own data in a CSV!  I'll keep shopping.  Let me know if you fix these issues.

June 28, 2013 04:54 PM
User photo
Luke Burns

I'd like to note I did the same thing as Alex. I have the same concerns. Waiting on an API and some more information on the accuracy of heart measurements.

June 29, 2013 11:27 AM
User photo
Martybacon

Last week I was sent a satisfaction survey and I gave my Basis Band a neutral rating because the info is difficult to save and share and the metrics are erratic and unreliable

but not a waste of money and effort as Basis is on the right track and future versions ought to be really worthwhile. 

Three days ago the display and interface began dying and last night it died completely. 

RMA can only occur if returned in the original box. 

 I am like most folk and discard packaging once the product has been seen to function properly.  Had I kept the original package, and gotten an RMA permit, the return would likely be a  
refurbished Basis Band, so that is clearly just a waste of time and money. 

My band worked for about a hundred days.  2 dollars a day is not too much to pay if the data were reliable and exportable.


Let me know when the Band is  improved and I will try again.

martybacon 

June 30, 2013 06:01 AM
User photo
Alex P
Marty,

I would say $2/day is ridiculous for a product like this. That's $60/mo which is more than many cellphone plans. That just for something that doesn't work very well. If this were $100 like the fitbit and fuel band, this might be worth purchasing but the fact that it's $100 more but still doesn't function very well is definitely a deal breaker. Not to mention, to all of the people saying they're on the right path, how do you know that when (or if) they ever get an up-to-par product, they're going to support you still? You bought this product if they build a new more working one they're likely not just going to send it to you.
June 30, 2013 07:41 AM
User photo
Eugene Bogorad

Come on, people! It's still the best sleep tracker as-is, without any API. I'm loving it and a lot of my friends are as well.

June 30, 2013 07:46 AM
User photo
Letmein

Marty, that is the first I have heard of needing to have the original box. In my experience, these people are looking after any issues with the product by replacing it with new product. Sorry about your experience. If I were you I would try and escalate the issue.

I have had 2 bands replaced due to issues like yours, and in both cases they sent me new product that I simply swapped out and sent back in the box they shipped it in.

They were nothing but extremely helpful in both cases so I don't get why they won't do your return as you are within the year….

June 30, 2013 09:07 AM
User photo
Lee

I want to second that I have not had a problem with needing the original box. My first band was swapped very easily they just asked me to mail the one one back after I got the new one. The swap went very smoothly.

June 30, 2013 09:31 AM
User photo
Martybacon

Thanks to you two for recounting your good experiences in RMAing to Basis.

 

I was going by what the web agreement  states, no one has yet gotten back to me from Basis so I dont know if they will try to play stricter with me than with you:

 For all returns, including warranty returns, you must pack your BASIS product in its original packaging and include all accessories and documentation.   We reserve the right to charge for any damage to the BASIS product, and missing part fees may apply.

Please contact Customer Service prior to returning any product to receive a return authorization form and RMA number. You will be responsible for, and pre-pay, all return shipping charges and shall assume all risk of loss or damage to product while in transit to us. We recommend that you use a traceable method of shipping for your protection.  We will pay for shipping to return any product to you.

Email us at support@mybasis.com to obtain an RMA number.  

June 30, 2013 10:26 AM
User photo
Letmein

Well Marty, don't you think you should give the company a shot at rectifying a problem for you before posting your dissatisfaction in the forums here. Like most new companies, their policies and return circumstances will most-likely adapt to what their user-base needs. I can't see why they would not treat you the same as us so I think you should let them know what your problem is. 

Apart from a couple of concerns I had regarding what appears will be a crippled iPhone App in comparison to the Android App, I really think these guys have done everything they can do to keep their customers happy.

 

Hope they are able to resolve your problem.

 

 

 

 

June 30, 2013 03:02 PM
User photo
Mike Fliss

Read the whole thread.  I've got a band - submitted an invite request about two months ago, but no luck.  Paid $300 on ebay for a used one, still working fine - which I'll probably sell and buy a new one when I finally get an invite.  

 

But I wanted to +1 the data dump request.  Users feeling like they own, or at least have access to, their data is huge.  I just ebayed my old zeo for this reason.  If Basis could stand up and say: "You can get your own data in CSV at any time off the website if you want to play with your own metrics.  And even if we go out of business (which we have no intentions to do), you'll still get to download your own data.  And as a parting gift, if that happens, you'll have iPhone and Android apps that can email you your data dump at any time so you can continue with our product."  That'd be a serious vote of confidence.

July 02, 2013 01:11 PM
User photo
Chris Smothers

All I want is access to my data. I just received my band after 3 years on the wait list and it's like having a vault of information on my wrist. The web UI is very pretty and surprisingly sparse. It seems like an odd use of time and resources. The watch already provides data that no other device can match. If there were an open SDK or at least an API there would be a  wealth of developers happy to build applications at no cost. With such an exciting technology and a community willing to help, I don't understand why Basis has decided to be so secretive about its plans. I would love to see Basis adopt a policy of full transparency like https://www.balancedpayments.com/ . They are constantly available to talk with early customers and work together with developers to build their API and plugins. I know I would be happy to test an unfinished API and give feedback. But over the past 3 years my several interactions with this company have been anything but transparent and humble. So while there are plenty of people hoping this product and company will succeed, I fear that you will not take advantage of those resources and this budding ecosystem will suffer.

July 02, 2013 01:56 PM
User photo
David Weddle

First let me say I'm digging my B1 so far.  Still, the more I think about this topic, the more baffled I become.

I've read most of the comments on the API and am left with the question; why is it so hard?  The API already exists, otherwise the system wouldn't work.  Similarly, there are job descriptions on the site describing a role of Director to own the RESTful services tier.  There are too many clear signs indicating that it isn't the creation of the API that's the problem.  Documentation, while not necessarily trivial, shouldn't be the obstacle.  Fear of a slew of support issues, that might cause the hold-up.  An active community, seemingly budding even on this thread, could help overcome that.

I appreciate the need to prioritize.  I'm sure resources are limited, but the decision is still curious (akin to printing and shipping the "Designed for iPhone" logo on the packaging before releasing the iPhone app) and difficult to fully understand.  

@Kristin, maybe a bit more insight into why would help.  At least it would help me square Basis' timing of publicly available API with  its policy of "All Biometric Data shall remain the sole and exclusive property of BASIS Science, Inc."

July 02, 2013 06:28 PM
User photo
Damien Blenkinsopp

I'd like to get my data from Basis into tictrac.com (allows comparison of many metrics)- with some of the hacks in github etc that have been brought up, has anyone achieved this?

July 06, 2013 04:54 AM
User photo
Brendan Paradis

It would be a real shame if we had to resort to third party application connectors to retrieve our data in order to have access to it directly. We need this!

July 08, 2013 09:32 PM
User photo
Alex Finlayson

Having an API/Export data is more important to me than an iPhone app. It's really unusual that Android got an app first especially since most people in the QS area are iPhone users. I also believe iPhone users are more likely to buy new hardware and are more likely to be first adopters. Hell even if there was just a protocol for the bluetooth connectivity, I could make my own app.

 

You said that your problem is that you are a startup and don't have the resources ... WRONG! you are ignoring your number one resource you have your first adopters. If you open it up WE become a resource for you. 

If you need help alpha/beta testing APIs send me a line, I'd love to help.

Alex

 

 

July 10, 2013 11:36 AM
User photo
Gioel Molinari

need this badly

July 11, 2013 09:47 PM
User photo
Joe Rosenblum

+1 this would be great (though I would put this lower priority than the iPhone app)

July 12, 2013 01:56 PM
User photo
Yuri F M Malina

Definitely need this as well. ASAP +1

July 14, 2013 07:28 PM
User photo
Nathanael Greene

I have played around with the scripts to get the json files. Getting the files of data proved not too hard, but json files are not very helpful for someone with pretty modest tech skills such as myself. Anyone written a script to convert these json files into csv or excel? 

Also anyone who has extracted their data found out anything interesting from it? Given that the band is extremely inaccurate in an absolute sense and really more useful for identifying patterns, the habits and insights are a remarkably poor use of the data. Rather than setting absolute count thresholds, after a week or two, the insights really should shift to pattern identification. (The "pattern" charts are of no use if you're not looking for a pattern related to 1 variable and the hour.) I've been wearing mine regularly since early January. That's offers the potential for big data type pattern analysis. I would love to hear what patterns if any folks have identified or tried to identify.

July 19, 2013 12:57 PM
User photo
Beau Gunderson

@Nathanael: There are JSON to CSV converters online, like these:

http://json-csv.com/

http://www.danmandle.com/blog/json-to-csv-conversion-utility/

Let us know if they work. :)

July 19, 2013 02:25 PM
User photo
Luke Burns
July 19, 2013 06:21 PM
User photo
Nathanael Greene

dunn5's save_metrics_as_CSV is really close. Of course you have to have or download python 2.7. (The script doesn't work with 3.3 or at least not seamlessly.) The output isn't exactly how I would structure it, but I'm going to see if I can get rearrange it in python or if not there, manually in excel. Then the question is given the huge dataset I'll have. Thoughts? I'll post what I can figure out.

July 22, 2013 09:04 AM
User photo
Nathanael Greene

Sorry... The question is given the huge dataset I'll have, what patterns should I look for? I've been having a hard time hitting my calorie goal, so I'm thinking of looking for the step, heart rate and perspiration patterns  that mark the days that I hit my goal and then seeing if there are patterns early in the day that correlate with achieving the component goals. Obviously correlation does not equal causation but it seems to me that I could give myself some interesting coaching tips based on this type of thinking.

July 22, 2013 09:12 AM
User photo
Davis Meanie

I seriously can't believe the basis still has no API. It's nice to be able to grab JSON, but that relies on the Basis site and whether it stays up. Sure, Basis is a startup and has limited time and resources so on and so forth, but the persistent failure to even provide CSV exports from the site (let alone from the device, which is the right way to do it) is incredibly weak. The product was demoed at CES 2012 - how long can CSV export actually take? Two days? A week, if the developer were righteously drunk the whole time?

Beau Gunderson: I wonder if the band is vulnerable to the null prefix SSL attack that Moxie Marlinspike found (and implemented in the SSLSniff tool). It would require getting an SSL cert though, which costs...

July 31, 2013 04:40 AM
User photo
Beau Gunderson

@Davis, I was able to sniff the USB traffic using USBPcap on Windows, it's viewable using Wireshark 1.10.1 and the captured data is available here:

https://github.com/beaugunderson/re-basis

There are two large packets that look like good candidates for the raw data from the band, I've marked them in the file.

July 31, 2013 12:07 PM
User photo
Karl Welch

This looks like and awesome device but without the ability to export a GPX file, all the wonderful data collected is useless to me. Even with fancy apps, I still can't submit my data for credit against my health plan so I would be forced to use another device. Personally I think Basis should consider that while they may be basing their priorities on customer request, one of the main reasons many people may choose not to become a customer is lack of access to the data, after all, at the end of the day that is the devices core function... to gather data.

July 31, 2013 06:56 PM
User photo
Jeron

All of the types of exercise I do, that I want to use this device to monitor are "low impact" types of exercise like bike riding and an elliptical machine. The BASIS can't properly record Steps for those exercises. Unfortunately, the Steps have too much influence on their algorithms for recording calories, so the feedback from the device is useless for me. (What does the device "think" is going on when I get my heart rate up to the same point it would be if I were running for an hour and it only records 50 steps? That makes me think of another wish list item, like a red flag when it senses "anomalies" that are potentially dangerous like a racing heart for no apparent reason, but that is for another thread...)

I need a data export, so I can use other algorithms that don't rely on Steps.

Until I get an API or data export, the only value the BASIS has for me is recording my sleep habits.

It can't be difficult to provide a data export. If it is a difficult task after all of the code they have written so far, then they have far more serious problems that need to be addressed...

August 01, 2013 09:13 PM
User photo
Sean O'Connor

This is 9 months now and still no open API or export?

This company needs to respond to this, yet again, and start to talk about a time table.  You're losing customers left and right at this point.

This isn't the only issue you're failing to address properly, so saying you're dealing with X, Y and Z first is no longer an acceptable excuse.

August 05, 2013 07:46 AM
User photo
Nathanael Greene

A Few thoughts and suggestions: 

  • While I get it that there are a lot of balls to juggle for a start up, the lack of an export makes me nervous that Basis has plans to sell me my data or, worse, sell others my data. So a simple reassurance the the data collected and currently stored on the Basis website is fundamentally mine and will eventually be available to me for free would go a long way to making the wait for an export function less worrisome. Upselling additional analysis of the data is fine, but if the company claims ownership of the basic data, that would be just evil. (Think Instagram.)
  • The lack of an export also makes me worry that there might be something wrong with the data. I haven't found any simple assessment of it's accuracy or precision and without an export its hard for the users to check it ourselves. More transparency on this would help.
  • Three specific suggestions on presenting the data: 1) on the patterns page, either show the cumulative total for steps and calories for each day or at least highlight the days that my step and calorie goals were met. It's hard to know what a successful pattern looks like without that data showing on the same page. 2) on the patterns, allow the user to set the number of bins for the variable being shown and to choose between 15, 30 and 60 min intervals. 3) For calories, autofill the time when I'm not wearing the watch with some sort of minimum or average. On days when I have to charge for a while or forget to put it back on, the calorie count is useless.
August 05, 2013 09:02 AM
User photo
Damien Blenkinsopp

Here's an idea - perhaps the quality of the data isn't good enough and creating an export interface would show this up. So I don't think there will be an export interface until / If and when that gets fixed. That depends on whether it is a hardware or software issue - it's most likely a hardware limitation IMHO.

I check in on my watch often and there are big inconsistencies, times where it gets stuck at a heart rate, times where it's way off (e.g. 160 vs. manual check shows is more like 60 regular heart rate, or the other way around), you've probably seen the reviews of Basis compared to other heart rate straps and watches etc. which show how far it gets off track when you're moving around. 

August 05, 2013 09:40 AM
User photo
Kristin
BASIS

Data export is a high priority and on our road map. As mentioned previously, we have limited resources because we are a start up. I wish we could accommodate you immediately. We'll be sure to update this forum when we have more information to share on this topic. Thanks for bearing with us.

August 05, 2013 10:06 AM
User photo
Gary Teal

Wow, Kristin, I hate to pile on, but I sincerely want my data as much as everyone else. You are coming close to failing the Turing test with these information-free posts. I suspect, actually, that you'd love to provide more information but you can't get your hands on it, either, so I don't mean this to be personal. Of course it's on your road map - but can you tell us where, even if not when? When you repeat that you have limited resources you risk scaring us all away. There's a ton of venture money out there if that's the only problem you have. We already assume you wish you could accommodate us immediately and that you'll update the form when you can, and we have no choice but to bear with you after we've paid for the watch. Lastly, I want to second the motion that you consider crowd sourcing this little task; it sounds from the thread that loyal customers are hacking out a solution already. Send one of these people who have already written some code a non-disclosure and they write a utility, probably at little or no more cost than giving them a bit of public credit for being a hero. Sorry if I sound mean. I want this product to be successful.

August 06, 2013 04:19 PM
User photo
David Weddle

I've posted another toy app https://github.com/absurdpoet/BasisData that pulls the Basis data from the website and loads it into a (cough, cough) Access database.  There is the beginnings of a CSV renderer as well.  Before continuing, let me state for the record:

First, I realize this project is, at least in part, duplicative of work already posted here.  Thanks to all those above who started the Basis JSON discovery.

Second, I kinda hope this is project is made largely irrelevant in the very short-term by Basis surprising me with its own approach.

Third, my coding skills are rather rusty.  There's a good bit here that could be done better/more efficiently/with more concision/etc.

Fourth, this is my first proper foray into open source.  Well, after you look at the code, you might want to debate me on the "proper" aspect.

Beyond the use of a (cough) database, the major difference is that I try to grab both the notification set as well as the observations.  I've placed a sample db in the data folder of the project for review.  The data set has been limited to a sample of the info from my B1.  The next steps, hooking up to R/Stata/SAS/your favorite analysis software, are straight-forward.  Fixing missing data, an idea presented in a previous post, is also pretty straight-forward (I'm debating whether I want to use previously observed value, some sort of rolling average, or a decay function).  ALL sorts of good stuff are possible.

You'll see the docs are missing.  Candidly, I'd rather not write them as I'd prefer to get the solution directly from Basis, however, since that timeline isn't clear, I'm willing to write them (sometime in the next couple of weeks) if there's folks out there that would be able to use this for themselves.  

The real reason I am pushing this out there without proper docs is that I want to demonstrate to Basis that:

One, there is real interest out here.

Two, communicate that, for all its potential protests, we're going to start hooking up to the pseudo-APIs that are currently available.

Three, determine if there are a few others that would like to use this solution (i.e., give me a real reason to write the docs).

And finally, let me close by saying, again in list form:

One, Basis, while I doubt you'll want to, please consider publishing your algorithms to determine what is sleep, what is a sleep interruption, what is exercise, etc.

Two. no doubt the calorie algo shoots low.  While I can adjust it in my solution, that's not the most user friendly solution.

Three, please please please please please provide us access to the accelerometer data.

Four, I'm sure I'll think of something else.

August 06, 2013 07:13 PM
User photo
Sue Peters

Contemplating my recently purchased unit, given there is still no time commitment on access to the data.

August 07, 2013 06:39 AM
User photo
Shaun Sommer

I for one am hoping they put a very robust API so that someone who is enterprising enough can build a Windows Phone 8 app that syncs. I can pair my Basis with my phone (they see each other and pair up) but with no app to receive the data it doesn't help much.

August 08, 2013 07:21 AM
User photo
Roger Braunstein

Still waiting... only reason I bought my Basis (and reserved a spot on the list over a year ago!) was to program for it - please guys, let us use the device.

August 09, 2013 02:01 PM
User photo
JS

Creating API with associated T&C is not easy. Especially since Basis owns all of your data right now. If you start exporting it OR not uploading it to Basis this limits their potential to monetize and create fun features. According to Basis privacy policy they own our data and they can do whatever they want with it (http://www.mybasis.com/en/legal/privacy/ - ex .- we may share or sell aggregated, de-identified Biometric Data for marketing/sales).

August 09, 2013 11:05 PM
User photo
Lee

Thanks for pointing this out JS

The line in there that most concerns me is

"When your subscription ends your contact information may be retained so that you can re-activate the service with your previously recorded data intact."

If Basis switches to a subscription model I'm out. I'm concerned that this potential plan may be why we don't have a way to get data off without going through the website.

 

August 10, 2013 05:11 AM
User photo
Shaun Sommer

agree, if they go subscription I am out. A premium OPTION would be fine, but the basic stuff should be free just like fitbit, etc. For the price of this watch they better not charge.

August 10, 2013 07:02 AM
User photo
David Weddle

Accelerometer data, please?  PLEASE?  PUHLEEZE?  I'd really like it, if possible.

August 17, 2013 01:46 PM
User photo
Jonathon

Show trust in yourselves Basis if you want the public to trust you. People will stay with you if you have the best product. Open data!

Also desperately hanging out for realtime accelerometer data via bluetooth and a Basis iphone app that supports third party apps via an inbuilt web server (node.js for ios?)

August 19, 2013 10:01 PM
User photo
Jonathon

I'm a little worried that all of these delays are related to the patent disputes from last year.

August 19, 2013 10:22 PM
User photo
Damien Blenkinsopp

Check it out, basis exports aggregate data for a client: http://www.fourhourworkweek.com/blog/2013/08/20/soylent/  

Work your way half way down the page to the part on Basis with this link: https://dl.dropboxusercontent.com/u/193116641/Soylent/Snow.xlsx

So you could email Bharat Vasan to get your data.

August 21, 2013 06:54 AM
User photo
Sue Peters

thanks for posting that Damien!  I checked out the "data dump" from Bharat (second link you provided), and it is a summary of the data, not the raw data, unfortunately.  It's nice that they could provide the summary data, but I think what most of us are interested in is the actual per mintue/second data, so that we can perform our own analysis, etc.. maybe I'm wrong though. At least that is what I am looking for!

August 21, 2013 07:24 AM
User photo
Ivan
+1. Let me hack with me own data!
August 26, 2013 06:51 AM
User photo
Brian Sanders
+1 on basic raw data csv export. Don't know why anybody involved in QS products has to be even asked for this these days. Come to think of it, probably should be an ISO standard.
August 27, 2013 11:48 AM
User photo
Jeron

At the very least, just give me a way to export the data on the Pattern page. That page is the closest they come to giving me the data I want to see and they ruin it by hiding the specific numbers with a color range. The numbers are there if you hover over the color square. That CAN'T be hard to do.

August 27, 2013 04:49 PM
User photo
John Brock

I've created a Python module to retrieve your data from the Basis website: https://github.com/JohnHBrock/bandsaw

August 28, 2013 07:28 PM
User photo
Lance Higgins

I know this is probably of limited use but I have noticed the json being stored on my phone as well.

File Location:  /sdcard/Download/response.json

 

I've used tasker to grab the file, append time stamp to the file name and save it off.  I like the idea of being able to view all of my data locally on phone.  I've not gone through and made the data very user friendly but I'll probably look at that soon.

 

Even though it may not be relevant here is phone model, OS, and basis sync app versions.

Model:  SGH-I317 (Galaxy Note 2, ATT)

OS: Android 4.1.2 (Build MC3)

*Basis App Version:  v1.1.0

* I noticed that there was a response.json with a timestamp of the 3rd earlier but did not get around to looking at it until sync had run again.  My point is I do not think the presence of the file would be tied to only v1.1.0 but at least one prior version as well.

 

September 04, 2013 03:12 PM
User photo
Beau Gunderson

Excellent find Lance!

September 04, 2013 03:34 PM
User photo
Beau Gunderson

Also, if you're open to sharing your data do you mind throwing it up on a GitHub gist so we can see the format?

September 04, 2013 03:35 PM
User photo
Cici Green

Wish I'd read this thread before my purchase today.  I view this device as a fantastic set of sensors and I want to analyze my own data - as all of you do.  Perhaps it is their $ business model that locks us out of our own data.  So frustrating.

September 04, 2013 07:03 PM
User photo
Oren Goldberg

I just found out about this resource you might be interested in. http://openmhealth.org/
"Open mHealth is non-profit startup building open software architecture to break down the barriers in mobile health to integration among mHealth solutions and unlock the potential for mHealth. Through a shared set of open APIs, both open and proprietary software modules, applications and data can be ‘mixed and matched’, and more meaningful insights derived through reusable data processing and visualization modules."

September 05, 2013 09:01 AM
User photo
Anton Matterson

@Beau Gunderson

here is a gist of my response.json https://gist.github.com/maneti/6481605

also, for anyone interested in working with the data, I've created an android app which will download the data from the watch, its still alpha/beta, but might be useful for some people... https://github.com/maneti/LibreBasis 

 

September 07, 2013 08:56 PM
User photo
David Weddle

Does anyone know if Basis is still staffed or even funded?  The silence that emanates from them becomes more concerning every day.  The website is up, sure, but considering the target of "dashboard improvements" was on the list back in May and the site hasn't changed once in form since then leaves me very concerned.  Yes, I see signs of A/B testing in the HTTP headers, but I don't see signs of change in the user experience.  I'll also note the iPhone app also came out, but that isn't a huge software design/build challenge for most development firms.  Nothing ever showed up online from the SF meetup.  Facebook posts come in drips and drabs; the blog even less frequently.

Given the "it can be good to fail" meme (yes, this article, http://www.wired.com/opinion/2013/09/why-do-research-when-you-can-fail-fast-pivot-and-act-out-other-popular-startup-cliches/, prompted this thought), might we be witnessing it right now?

September 13, 2013 08:47 AM
User photo
Stellar

Last I heard they had a massive backorder. But if someone knew the Apple Roadmap accurately enough, it could prompted a 'pull the plug on this.' Alternatively, they do have a really strong technology platform. They could be preparing for acquisition.

September 21, 2013 12:58 AM
User photo
Lee

I'm feeling very concerned about the future of Basis. According to linked in Kristin no longer works there. She was the only one posting much of anything on this board. It seems like a bad sign since I can't find anyone who may have replaced her role.

September 22, 2013 08:50 AM
User photo
JSON-CSV Support

Hi guys, some good news.  I am the owner of <a href="json-csv.com">json-csv.com</a>.   You can now paste your BASIS data (in JSON format) into our website which will allow you to instantly download all your data in a spreadsheet for easy viewing.  There is a small charge for the conversion but  I think it is well worth it.  You will be able to see a preview for free.

Send me a message through my website if you have any ideas on how I can make things easier for you.  For example, perhaps I could write a new feature to allow your data to be emailed to you automatically as a spreadsheet every week.  Please let me know.

September 24, 2013 09:49 PM
User photo
JSON-CSV Support

Sorry, that link didn't display correctly.  The website is http://json-csv.com

September 24, 2013 09:50 PM
User photo
Sue Peters

HI JSON-CSV - do you know if it is possible to access the JSON data file from the iPhone?

October 10, 2013 12:29 PM
User photo
David Spatholt

https://twitter.com/mybasis/statuses/390181392933732352

Saw this on twitter yesterday: @mybasis: "@fredkelly [API] in the works but no slated date at this point. Definitely next year."

October 16, 2013 05:20 AM
User photo
Lee

From Basis next year means "far enough away that we hope you forget this time estimate a year from now"

October 16, 2013 05:28 AM
User photo
Mark McCarron

The lack of an API borders on stupidity.  Who creates apps and forgets an API???  The staff at Basis need to understand how to leverage community support as this reduces costs to a startup.  Create the API and let everyone else develop apps, or integrate the API into existing apps.  The broadens the market that you are tapping into and creates a rich and diverse software eco-system.  I really don't see what the big problem is either, its not like creating the API for this service would take one solid coder a good weekend.

October 16, 2013 08:28 AM
User photo
Kyle Smaagard

Mark, I don't think they forgot it.  My opinion is that they are creating a closed system where they can benefit financially from the collected data.

I did sell my Basis band a few weeks back on eBay (it actually held its value fairly well)...I hate closed systems

October 16, 2013 05:22 PM