 I've got 5 p.m. So I think we'll get started if you want to you can pull up our slides here it'll be at the bottom of the slides to I'm Jason and this is my colleague Brett and we're going to talk about sunsetting so why sunset sunsetting simply a means to an end not the goal itself so we want to start with talking about what we're really trying to achieve through sunsetting and our experience and talking from others we found that there's a particular set of challenges that libraries are dealing with that sunsetting can help with and we'll focus on just a couple challenges here academic libraries and our universities are changing quickly libraries need to change quickly too and there's lots of hurdles to making the kinds of transformative changes we need one of the big challenges with the changes happening now is that we don't always know exactly what changes need to be made and when you're in that type of situation one of the worst compounding variables is being too busy we're all already busy at the same time that we're being asked to do more indifferent people need time to discover what the future library could look like but our people are already busy and that's often at the heart of an inability to change fast enough and I only know two solutions for being too busy you either hire more people or you do less if you know of others please let me know and we'll have discussion period at the end so you can fill me in on other solutions for busy this is a talk about doing less lots of ways to do less you can work smarter gain efficiencies apply automation you can say no you can be more selective what you take on lots of different ways to do less but what less looks like to us is staff having slack time slack times needed for innovation we need at least some people to have some proportion of available unscheduled unmanaged time busyness fills up time and excludes opportunity time this is a quote from Tom DeMarco slack what I call bankruptcy of inventiveness is often the result of a failure to set aside the resources necessary to let invention happen the principal resource needed for invention is slack when companies can't invent it's usually because their people are too busy busyness and effectiveness aren't the same thing so to restate the core problem is that just as we're trying to make changes we and our colleagues are too busy and we need to create space for new work in order to make the innovative changes we need to in strategic directions so we've come to thinking about sunsetting through thinking about how to do less in order to increase the amount of slack time we have to create new services and one of the advantages over other approaches is that we found that sunsetting has the advantage of maximizing the amount of recovered time so before I get into some specifics I want to say a few words about our background and the perspective we're coming with to this work we both work in the digital library initiatives department at NCSU libraries this is one of the libraries IT departments we're department responsible for creating novel services and applications using a variety of technologies our department has had its fair share of successes and as a result we've begun to encounter the limits of what we can sustain so many of our solutions to creating space are focused on technical projects and products but we believe that these approaches are broadly applicable okay now we've talked some about why you may want to sunset a service let's get more into the nuts and bolts we don't want to have like continual open questions about the future of a service that can really take a tool on staff so we like to save ourselves by setting trigger points for when we might consider a new direction and while only some of these might end in a sunsetting their useful moments to pause and consider that as an option a lot of our work on sunsetting developed when one of our library and developers left we had to take this large portfolio of successful products covering a wide variety of types of projects and technologies and distribute them across the department it really hit home how we needed to take these kinds of events much more seriously if we were going to keep moving forward and have the time to do that and we want to have the opportunity to address some of these questions before someone leaves so we've begun to identify some other trigger points we know that a staff change including an arrival is one trigger point for considering changes including sunsetting you'll need to pick your own trigger points but here are some suggestions for trigger points from the kinds of technology work that we do if you have to touch or change a large code base in a significant way then consider options like sunsetting is it worth it to put in the work needed to keep the project working there's a lot more to say about the importance of maintenance and the need to better valorize this necessary work but in the lifecycle of technology projects maintenance often exposes a particular organizational challenge that we've seen the true cost of maintenance is often hidden and why is that we think in part at least it's because technology projects start with an even distribution of effort amongst various stakeholders and technical staff but once a product is in maintenance mode it can sometimes lead to an uneven distribution of effort there's a long tail to these sorts of maintenance costs even if the percent of time devoted to each product might be low over time each still takes up some headspace and can lead to more anxiety and contests context switching costs another trigger point can be changing skills we found that over time we've made better choices on maintainable software stacks improve some of our development practices over time we've consolidated more around certain technologies that are meant to get the job done and this has meant that we're no longer developing and updating skills in some technologies that we used to know and for a department of our size maintaining depth and redundancy of technology skills across the department is is a real challenge we'll also see that skills in the market change for instance maybe you are a PHP shop and you're finding that you're hiring more Python developers consider sunsetting when you're not actively cultivating the skills needed to continue to maintain a service and since the time you've developed an application have commercial options become available matured at that point there's often no longer compelling reason to maintain a bespoke application you might need to live with something less bespoke but it can be really worth it for the amount of save time that you can recover so there's lots of reasons why organizational change might trigger sunsetting ones when you see the strategic direction and goals of your university changing you get this explicit statement of a pending change if you want to realign quickly consider what needs to be deep prioritized and potentially sunset okay so you've hit a trigger point that sets things in motion but that doesn't really determine the value of something and I can't answer for you what you'll find to be valuable so you you'll need to evaluate impact by your own metrics we do have one suggestion though on where you could start one way that can be that we found that can be effective for determining value is asking the question what if we sunset this starting the conversation there really presses every product and service to argue its value and new within the current context even if it was once successful is it still again this is a reason to stick to your trigger points you don't want to fatigue folks by continually calling everything into question excuse me so you've made the decision to sunset how do you proceed well sunsetting work is first and foremost a project and the first thing you'll need is a team but this team can be different from the original team who created the project being sunset we found that starting a sunsetting project often immediately freeze up some time for the highest performers on our teams who are carrying the bulk of the maintenance burden making sunsetting a project also privileges this work and helps give credit to folks who make this kind of work happen the second thing you will need is a communication plan our background is in technical products and projects but we found that what needs the most lead time in sunsetting projects is the development of the communication plan part of this is internal communication for the affected teams departments or the library as a whole the projects that get sunset are often someone's baby and they reflect significant achievements in people's careers it can be difficult to let go of things that have been part of someone's professional identity and sensitivity to that is important for internal communication plans the other part of communication is addressing the challenges around external audiences users faculty campus partners etc these may all be impacted by sunsetting in some way what options exist for data export or migration to other services and what services do you provide for them moving forward part of this work should be the exploration of alternative options if any for a user to have the remaining needs met so I'm going to discuss some of the paths we've taken to help create more space for new work and share examples of projects we've been able to sunset at NCSU libraries however as Jason said it's important to note that this is where the values of your own organization come into play and different approaches and strategies might be needed in your own context I'll also share alternative approaches that are on the sunsetting spectrum we've tried to focus on sunsetting as the default choice when evaluating mature projects but there are other near decommissionings that you can also consider if you can't get the traction for sunsetting outright sunsetting has the advantage of maximizing recovered time over the other approaches and that's the primary goal of this work to greatly increase the excuse me greatly decrease the amount of effort being put into mature projects in order to create opportunities for new work so don't laugh we maintain a wiki called wolf wikis for way too long where the NC State Wolfpack that's where that comes from this is obviously a mature area with commercial options available but people were still using this particular implementation in order to sunset the service we developed a communication plan for users and provided options for how they could export and use their data we learned that it often takes some research to learn her who your current users actually are and this took quite some time to sunset especially due to the large amount of communication with faculty before you communicate with your users you'll need to develop some migration options we don't want to just abandon our users outright how do we still direct our users to services that will meet their needs we ended up not needing these for wolf wikis but they were a good safety net to have available you may be noticing a theme and the naming strategy another service we recently sunset is Wolf walk a photographic guide to the history of North Carolina State University optimized for mobile devices this was NCSU libraries first iOS app and a testbed for GIS and mobile technologies we lost our in-house expertise in iOS development so we made the decision to sunset the project and this is an example of how a staff departure and skills changes can be a trigger point for sunset another example group finder was an application for helping students meet up in groups of the library turns out students now use social media or SMS to coordinate use of spaces thing called Twitter this is an example of how the commercial trigger point in this case text messaging can provide a viable alternative and allow for sunsetting of the service group finder came out before text messaging was such such a thing so changing environment another important note to consider is that sunsetting has nothing to do with the quality of work that was done for example lentil was a success successful project that played a big role in the early publicity for the James behind junior library it jump-started part of our social media archiving program and also provided a testbed for some large-scale displays and hunt sometimes however sunsetting is out of your control due to recent changes in the Instagram API thank you Facebook we are in the process of developing a sunsetting plan in many ways Instagram has made the decision about whether this project lives on or not so beyond sunsetting another general approach you can take or migrations and migrations can be a way for projects to live on but in different forms especially ones that take less effort to maintain some migration paths can be used to breathe a sustaining breath into old projects and some can be used to demonstrate that a project really is ready to be sunset we won't cover the whole spectrum of alternatives but here are some ways you can reduce the load of services even if you roll out outright sunsetting building on some of our development practices we've experimented using vagrant development environments as a migration target these are isolated disposable environments that can emulate a web server in every detail that are run on a local machine enter collections views the aim of collection views was to use data to help us understand how our collection expenditures relate to different departments at NC State the service was no longer heavily used but still has potential future value thanks to the depth of expertise held by the stakeholders a virtual environment was a viable way to both remove the service from production but to sustain it in a way that we could resurrect it if we need to do static sites or another migration path and are one way to keep an application alive with lower effort and higher sustainability oftentimes by the time a project reaches maturity we have a much better understanding of what features or capabilities are most important we can take just the most useful important parts of an application to allow it to live on this isn't a sunsetting strictly speaking but it's also maybe a point that you don't get to without raising the question of sunsetting it's also worth noting that a static site migration can be completed by someone other than the original developer say a student one static site migration we've done is red white and black which highlights African-American history at NC State in combination with an annual tour led by NC State faculty this is obviously important historical content but since it doesn't change much over time it was a good candidate for static site migration it also demonstrated how static sites can preserve core features of original services static sites don't necessarily mean a loss of functionality for example search is now possible client side using JavaScript we can also use some of our existing infrastructure to smooth the path to sunsetting such as using web archiving to create a snapshot of a site we did just this with the learning space toolkit a grant project deliverable that we migrated using web archiving tools it was really great to see Dr. Joan Lippincott reference this project during her talk this morning and hopefully this shows how the web archiving static site approach can extend the useful life of a project while also bringing closure to a major grant deliverable those are just a few of many recent examples of sunsetting work we've done at NCSU in addition to team formation and communication planning each of these approaches have some common actions that should be taken to ensure the success of sunsetting work first we need some ways to capture the history and success of projects so that people can continue to be recognized for the great work that they've done we also need to preserve what we've learned from a project for internal needs one way we do this is through project pages that capture who worked on the projects describes some of what the project was about and shares some of what we think the major impacts were and it also serves as a way for external folks to get a historical view of our projects we also need a way to capture the parts of the project that are essential for the organization the internal part of the work this might be documentation about code repositories database backups or project documents finally perhaps the biggest lesson we've learned as a result of our recent experiences sunsetting projects is the importance of considering eventual exit paths from new projects during their conception rather than deferring these considerations to a time after a project is mature and has become maybe burdensome in some way okay so I'm gonna talk some about how we do portfolio management and how that fits in we start portfolio management with the individual staff person we occasionally ask staff to update their portfolio of current work this is a much lighter weight process than something like an annual report the audience is just managers in the department it's isn't seen any further can I steal have you seen any of these no okay and it's not a tool that we use in evaluations we keep it brief it's just a snapshot in time and we largely create and manage portfolios to identify trigger points and evaluate the amount of slack time that staff have okay so what do you include in an individual's portfolio we start with the products and services the person works on of course we want to gauge the current level of effort involved in their major work we list projects initiatives experiments so that we can identify places where we can fail fast what we effectively can sunset early sometimes we'll see an opportunity we want to take advantage of it and these are often places where there can be some give in schedules and all we list committee assignments sometimes one person can keep getting tapped for committee work reviewing and requesting changes to membership can recover some slack time for staff and of course you know committees can also be sunset if they're no longer useful and we include service and really anything that's going to take a significant amount of time for every product project committee etc we note the current status is the work still active or in some kind of maintenance mode we also asked staff for a recommendation for how to proceed what do they consider to be the current value of the work in their portfolio it's really important to get the perspective of those who are closest to the work as managers were encouraging folks having a job complex that includes variety we like to create a portfolio with a good mix of projects and products at different stages of development then we can also balance portfolios across the department this can help even out say responsibility for maintenance versus pursuing new ideas we can recognize of a trigger point has been hit and in some cases we can find products that can be sunset and we think that this approach to portfolio management can also hopefully feed up into higher level strategic planning as well the portfolio for the library okay so to sum up we want to be an innovative department in library where people have the space to do the new things that need done sunsetting is a way to maximize the amount of slack time recovered to allow us to better innovate and active portfolio management has helped us identify where we might be able to clear space for new work and if you'd like to read more about sunsetting we've begun writing up a short booklet it's a work in progress but you can already find some extras there that we couldn't include in the presentation today like why we choose the term sunsetting over other terms that we could have used and we'll be including more potential trigger points there examples and the like and we'll we're also looking at writing up more in-depth some of the case studies of successful and unsuccessful sun settings so we'd be interested in talking with you more and finally you can also find a link there of another slide deck that we did that has a Halloween theme to it if that's your thing um questions