With the launch of MapD version 3.2.3, we made a few changes to MapD Community Edition, and the download page for it on our website, that I think are worthy of a quick explanation. More details on the specific changes below, but I think it makes sense to start with the three goals we had in making these changes:
- Reduce the friction for first-time users to get their hands on Community Edition.
- Increase the diversity of platforms we target, and improve the installation guides for each of those platforms.
- Increase our understanding of how Community Edition is being used, so we can continue to improve the product.
Okay, let’s look at the specific changes.
An Open Download
The first change we made was to our download page on mapd.com: we removed the email gate. New users used to have to give us an email address, and then we would send them a link to download the Community Edition, but we’ve now replaced that with an open download page. No more hurdles, and no more marketing. Just choose your target platform and install method and get the detailed instructions for getting MapD up and running straight away.
Also note that we have specifically called out GPU +CPU and CPU only installation methods. We’re pretty proud of how well MapD runs on CPUs alone, when you need it to, so we’re happy to make it even easier for people to get MapD for CPUs.
Additional Download Targets
We’ve also expanded the number of installation targets, supporting two flavors of Linux with their respective install methods, and a simple container install with Docker. Each of these methods comes with step-by-step instructions and a video tutorial to walk you through the process.
Of course, if you run up against any problems, our Community Forum is always available to help.
Finally, we added some “phone home” callbacks into Community Edition that send anonymized information back to MapD about how the product is being used. We’ll be using this information to help us better design and engineer the product, and help answer some questions about where we should be prioritizing our development resources. We also updated the Community Edition EULA to reflect this change. We have always had a clause in the EULA (section 3.2) that talked about our collection of diagnostic data very broadly, and now we’ve updated that section to clarify what we’re specifically collecting and why.
We are of course not collecting any of the user’s data that is being uploaded into the database, or the specific SQL queries being run against the database. We do not want that data. We do collect the size of the database being queried, to help us better understand the dimensions of use, but that does not tell us anything about the data itself.
If anyone is concerned about our collection of this additional information for product improvement, there is always the 100% open source MapD Core and related open source APIs. We have not changed anything there. We are not, and will not, be adding any of this information collection at that level. Only for the Community Edition, where we’re distributing a complete, easier to install package including the non-commercial license on MapD Immerse, do we think it is fair (and industry standard) for us to ask for this small amount of anonymous information in return.
If you have any questions about any of these changes, please respond here or reach out to me directly (firstname.lastname@example.org).