Increasingly our clients are seeking ways to share data both internally with other tools they use, as well as externally with one or more partners.
An example of connecting internal tools would be a client of ours connect their phone system with iCarol both to facilitate “screen pops” when a call is routed to a particular phone worker and iCarol appears prepopulated with information about that caller, as they answer the phone; as well as to combine the data collected by both systems to answer operational questions like “what is our average handling time for calls related to different help seeker needs?”.
And an example of sharing data externally would be giving access to your resource database so a third party can build a mobile app or a website targeted at a certain sub-population in your area, like immigrants or job seekers.
Enabling these data sharing relationships, whether internally or externally, is where API’s come to the rescue.
An Application Programming Interface (API) allows electronic systems to interact with each other without the need for the direct human intervention. That is, with an API no person needs to direct data traffic between two systems, say via a website or other screen – the systems just talk directly to each other behind the scenes. In this way, the data that resides in the main system can be searched, retrieved and even modified by other authorized computer systems connecting to it.
To do this, a software vendor writes an API and makes it available securely on the internet, and also publishes documentation about how other developers can use it. They can be one-way APIs, also known as “read only” because the software consuming the data from the API cannot modify it. Or they can be two-way APIs, where the consumer software can make modifications, like creating new records, or modifying or deleting existing ones.
At iCarol, we’ve had an API for a number of years now and are actively expanding its capabilities – it is used by quite a number of our clients to enable real-time data transfers both internally and externally. And we also consume quite a few API’s published by other software systems. Some of them enhance iCarol’s capabilities, like Google Maps or tools that let us send and receive SMS messages within iCarol. Others let us push client data to their partners, for example client or call data that needs to transfer into a partner’s electronic medical record systems.
APIs have been around for a long time in the software world, and will only grow in importance in the years to come. We continue to be excited about their possibilities and will certainly be expanding our use of them.
Being a web-based service, we are always aware of how access to our systems effect you. After all, many of our users operate as 24/7 services, entering call report forms and searching for resources ’round the clock. We recognize that there’s never a “good” time for iCarol to be unavailable to you, and this is why we’ve invested so heavily in infrastructure and processes that work hard to minimize outages and make unavoidable ones as brief as possible. We have “fail over” servers that are are set to pick up systems routed to them in the event of connection loss, and members of our technical team receive notifications no matter the time of day or night in the event of detected instability so they can begin working on a solution, often before our users have even detected a problem. In an effort to be completely transparent about iCarol’s uptime and availability, we make status.icarol.com available to you, and advise our users to check this site if they suspect they’re having connection issues to help determine whether the issue is originating with iCarol or something is amiss with their own local network connections. These are all examples of efforts and placement of reliability as our very top priority.
Our service level agreement with our users is 99.9%, which translates to about 10 minutes of allowable downtime, on average, per week to achieve. Of course it’s our goal to go above and beyond that, and consistently we do. iCarol regularly achieves uptimes of 99.95% or 99.97%, and our overall uptime for 2016 was over 99.97%, translating to no more than three minutes of downtime per week on average. This 2016 achievement exceeds industry standards as comparable to some of the top web-based service providers, including Microsoft Dynamics, Salesforce, Amazon, and Netsuite. These companies are, of course, much larger than iCarol in every way, and yet we often meet and exceed the service levels offered by them.
We’re pleased to say that thanks to some of the infrastructure work that’s been done over the last year, our total uptime for the first quarter of 2017 met a “four nines” achievement, which is 99.99% uptime. To give you an idea of how good that is, it means that in a given month a site would be unavailable for just over 4 minutes total, and about 1 minute of downtime weekly.
Occasional, brief slowdowns or short periods of downtime are unfortunately an unavoidable part of the online world in which we live, but making these interruptions a rarity is our top priority. We are scaling our infrastructure and systems as we grow so that we can continue to provide the level of service you’ve come to expect and deserve from us. It is an ongoing item that we’re always monitoring and looking to innovate, but we were particularly happy with the performance following the strides we made and wanted to share that with you.
In case you don’t know him, our CTO Neil loves to code. And he also loves to address social challenges with technology. So when the University of San Francisco’s School of Management announced their one-day Social Entrepreneurship and Innovation Conference and subsequent two-day Tech4Good Hackathon, he immediately signed up to make the short drive down to the event to see how he could help.
About 70 bright individuals arrived for the hackathon’s opening ceremony on Saturday morning and, one-by-one, introduced themselves to the group by announcing their skills, why they were there, and an idea for what problem they would like to see solved in ways that would have social impact.
In the ensuing hour, participants informally coalesced into the teams that would start building their entry into the hackathon’s competition, to be judged the next day by an esteemed panel of five high-tech executives from the Bay Area’s startup scene. By introducing himself as a “CTO, architect and full-stack developer” Neil was in high demand to join numerous of the 10 teams who hoped to have a useful, coded prototype to show about 24 hours later.
As he has a family member who could benefit from it, Neil settled on a team to build a web-based tool that replicates a “real world” tool that helps young learners with phonics, letter identification and word formation in their journey toward reading fluency. Each of his other three team members brought very useful ideas and skills and were amazed at how quickly they organized into a highly functional and productive group.
After more than a few energy drinks, late night coding sessions, and last minute changes, the team submitted their entry for judging on Sunday afternoon. Although Neil’s team didn’t win, he helped other teams throughout the weekend and was impressed with all of the final entries.
Although it was his first hackathon, we’re pretty sure it won’t be his last!
While we are proud of the fact that our total uptime in 2015 of 99.955% exceeded our goal of 99.95%, we understand that any downtime, no matter how brief or infrequent, impacts you and ultimately the people you serve. On February 28th we’ll complete a transition to new, even more powerful infrastructure to support our users. This change will affect our NA0 data center in Toronto upon which most of our clients are hosted, and will occur without impact or downtime to our users.
Once the transition has finished, you’ll enjoy enhanced system response that will improve your workflow. The redundancies included in this upgrade will drastically reduce already infrequent downtimes. This technology investment will also create an even more scalable system; as you build your partnerships and as we continue welcoming new helplines into the iCarol family, you can rest assured iCarol will continue to grow and will be able to more than accommodate your needs for added activity and data storage.
We look forward to continuing to serve you by aggressively investing in the best systems to support your life-saving work. If you have any questions about the upcoming upgrades or reliability, we invite you to open a case with our support team, and we’ll be delighted to assist you.