skip to main content

TAMU Webmaster's Blog

Information and insight from the A&M Webmasters

Stone Soup

April 27th, 2009 by Erick Beck

Over the weekend I read The Pragmatic Programmer, by Andrew Hunt and David Thomas.  One of its early chapters uses a story commonly known in eastern Europe to illustrate a point

The three soldiers returning home from war were hungry. When they saw the village ahead their spirits lifted—they were sure the villagers would give them a meal. But when they got there, they found the doors locked and the windows closed. After many years of war, the villagers were short of food, and hoarded what they had.

Undeterred, the soldiers boiled a pot of water and carefully placed three stones into it. The amazed villagers came out to watch.

“This is stone soup,” the soldiers explained. “Is that all you put in it?” asked the villagers. “Absolutely—although some say it tastes even better with a few carrots….” A villager ran off, returning in no time with a basket of carrots from his hoard.

A couple of minutes later, the villagers again asked “Is that it?”

“Well,” said the soldiers, “a couple of potatoes give it body.” Off ran another villager.

Over the next hour, the soldiers listed more ingredients that would enhance the soup: beef, leeks, salt, and herbs. Each time a different villager would run off to raid their personal stores.

Eventually they had produced a large pot of steaming soup. The soldiers removed the stones, and they sat down with the entire village to enjoy the first square meal any of them had eaten in months.

The soldiers here act as a catalyst, bringing the village together so they can jointly produce something that they couldn’t have done by themselves—a synergistic result. Eventually everyone wins.

How we handle IT on this campus really struck home while reading this story. We all know how decentralized things are, and we all face the prospects of reinventing the wheel because we all need many of the same projects. Yet we continue to do so because we can’t, or won’t work together. Like the townspeople in the story we keep our resources to ourselves rather than combining them together toward a larger purpose, even one that we can draw benefit from.

Since John and I have been here we’ve tried to begin the building of a quality university web presence, and we have lots more left on our plate to implement. But we can’t do it all, we have a staff of four and that’s barely enough to keep our heads above water. What we do have, though, are the stones to throw into the pot in order to get the soup started. Through collaboration we can do much more together than any of us could ever accomplish ourselves. Over the next few months we’ll be rolling out a few projects and will likely mention how “it would be much better if only we had…” I hope we can all enjoy the feast when that time comes.

Monday, April 27th, 2009 Future Projects, Uncategorized
Share this article

28 Comments to Stone Soup

  1. I just had a discussion with a friend about this notion of being over-committed. I think this stems from a lack of assigning a value to a project and prioritizing.

    I think the American notion of how to bail yourself out when you’ve got too much work is to work harder or to throw more people at the problem. Mythical Man Month and personal experience tell us this is not the case.

    The Lean solution to the problem of “we’ve got so much work that its a wonder that any of it gets done” is to look at what really adds value to the customer and or end user. I think the “customer” or “end user” in most of our cases is one of the three: the students, the A&M community, or the general public.

    I know that probably almost all of our projects have some level of value to someone in the above mentioned groups. The question is not which projects deliver value, but which projects deliver enough value to avoid the chopping block. We are all overloaded. I see people talking about this all the time.

    So the only question remains, what do we put first, quality or quantity? Should we concentrate our efforts that are sustaining and strengthening to the vision of our colleges and departments? Or should we try to serve as many people as possible regardless of the cost?

  2. Robert Stackhouse on April 27th, 2009
  3. Good thinking.. In the story, in fact, the villagers actually provide the pot, water, and stone! The soldiers didn’t have to do a thing.

    I would just caution against too many pots being set to boil: establish a central town square and equal opportunity to contribute to the soup and partake in its yumminess.

    svn vs git? 🙂

  4. bronius on April 28th, 2009
  5. I’d rather go with Bazaar. It works better on a Windows environment. Before anyone goes calling me “stupid and ugly” though, I’ve got to say I’m running Ubuntu Server on my VPS and the home office computer dual boots into openSUSE.

    I like the network of trust aspect of distributed version control.

    Setting up a main dev branch is dirt simple using Apache or even IIS.

  6. Robert Stackhouse on April 28th, 2009
  7. […] commented on a post the other day, and that post got me thinking about how to successfully bring about lasting change […]

  8. “You’re stupid and ugly” ain’t going to cut it. :: Robert Stackhouse on April 28th, 2009
  9. I agree with bronius on making sure we don’t start too many pots and like the idea of the central town square.

    I’ve read that book too, pretty good stuff. And Erick, I think instead of you just having the stones to throw into the pot to get it started, you already have the Google Search Appliance (GSA) too.

    I’ve thought a few times about trying to get people to tell what apps they are using/have made for different tasks, and not just the big ones, but the smaller stuff too. Then see if they could be reused, and if so if they could be used as a hosted app. I know with keeping track of costs we might have to pre-figure a cost for an app, to speed up paperwork between colleges/departments. Then other groups could use that app for the pre-set cost and put in their own look in some way like how the GSA works. We also have the issue of people seeing something then asking if they could have it too, which isn’t bad, it just creates multiple copies of the same thing and maintenance gets tougher. A hosted app would also take away the code-sharing problem of which language an app was written in.

    We could also set some specs on how to build an app that other people could use when you start one (or adjust one), such as enabling multiple instances, being able to configure the instances, and ability for each instance to have its own layout.

    I think it’d be interesting to see what people are actually running and if they’d be interested in some app-level web service type collaboration.

  10. Craig Thompson on April 28th, 2009
  11. I think there is a lot of value in providing templates, project starter kits like this one, or anything else to get people up and running with new projects quickly.

    I think some talk of best practices would be good as well: caching, zipping, CDNs, unit testing, continuous integration, etc.

    I think there is a lot of collective wisdom on how to build quality, thoroughly tested, usable, and scalable apps in this community. I just don’t think that knowledge is being shared well.

  12. Robert Stackhouse on April 28th, 2009
  13. Somebody else going through basically the same thing we are and taking a similar approach…

  14. Erick on April 29th, 2009
  15. Robert,

    Good thoughts. I completely agree that knowledge isn’t being shared well on this campus. Another thing is that we’re a LAMP shop (the P is mostly PHP, but some Perl too), and don’t run any Microsoft like it looks like you might. If we had people working together at the app level running different back-ends, that would be a good example of what could be done without working together on source code, too.

    And to all,

    Would it be a good idea for us to start getting together as web developers on campus and discussing these things? Everything from code to apps to projects and requirements. We could take turns talking about projects we’re actually working on too. There’s a UWeb group on campus, but from what I can tell, it’s voluntary and likes it that way, and I’ve never really heard anything out of it since I’ve been here.

  16. Craig Thompson on April 29th, 2009
  17. Craig,

    I’ve been wanting to see the A&M equivalent of Google’s Tech Talks happen here for quite some time. I think this is hardly out of reach as we are a major public university. My former supervisor has got contacts (working developers) in everything from Google to Microsoft. A friend of mine is working at Pixar. We’ve got the network to make this sort of thing happen. What it needs is leadership.

    I think that we should form a self selecting committee to try to create a forum where developers can learn and share their knowledge. I think those of us that were involved in the web accessibility initiative did pretty good things under Erick’s leadership. I don’t see why a few of us concerned citizens couldn’t do the same thing for the entire A&M tech community.

    I think there need to be more opportunities for professional development here. That’s why some friends and I started AgileBCS: That’s why I continue to support the Aggieland DNUG, RefreshBCS, and the BVAUG in whatever way I can. I think an unofficial or preferably official A&M presence in this sphere is sorely needed. Did you see the link Erick posted above? At that university, they have a leadership program for university employees. Does A&M have something like this? If so, where do I sign up?

    I am language agnostic. I will work in whatever language/framework/popsicle flavor makes sense. I run a VPS in Ubuntu for personal use. My blog is in WordPress. I really think we should talk about developer skills that transcend primary programming language, yet again, this is why we formed AgileBCS.

    Lastly, I’ll be giving presentations on leadership and community building in the near future, so Craig, I hope to see you there.

  18. Robert Stackhouse on April 29th, 2009
  19. I think all the major concerns have been aired here.
    We don’t have any centralization or cooperation among groups.

    I too would like to see collaboration like google’s tech talks or TED. We have a great diversity and I think we really need to share that to help the community flourish. It’s painful to realize how much waste the university accepts to get business done as usual. I’ve brought my concerns up many times, but have gotten the response, “that’s just how we do things”. To an extent, I understand some of the complexities such as whose budget code is going to fund this and what benefit does MY department get from it.

    But the amount of money we could save by collaborating would boggle the mind. Even something as simple as sharing site licenses for software. Imagine a department that buys 5 licenses of Visual Studio, then consider several departments actually do this. Now contemplate what would happen if we just got say a 200 or 300 seat site license for the university and shared it between departments. I’ve not run the numbers on this, but I imagine you would realize some pretty significant savings.

    I ask, why don’t we do this?

  20. Jae on April 29th, 2009
  21. Sounds good. I like the idea of bringing people in, too, for extra ideas and support.

    I’ve also worked in other languages, but that was just an example of code that’s already written here on campus. I think for something to take hold on campus we need to have examples of projects and apps that are being shared or could be shared, and there is support for them for others to be able to use to help get people going.

    Another thought – how about we make a list of web developers and what group you’re with and how to contact them on campus? Would this be something we start on UWeb? Or make something new? This could include what projects you run/are working on and what plans you might have, and much more.

    Lastly, the AgileBCS thing looks interesting. I also do think we should start a group unofficially on campus to see what interest there is, then depending on what we get, try to see if we could find a sponsor for it to make it official.

  22. Craig Thompson on April 29th, 2009
  23. It has been my experience that at lease a portion of the development shops/groups around campus are aware of this issue of decentralized units, but no one has stepped up to leading the revolution. I commend you, Erick and John, for your efforts thus far.

    There is a site,, targeted at helping IT professionals by consolidating useful resources and solutions to questions that will inevitably arise from staff members unfamiliar with TAMU. I think a similar situation will easily exacerbate the problem being discussed here when new talent arrives at the University and is quickly indoctrinated with this sense of a departmental silos of information and development. What Craig mentions above is a great start consolidating relevant information about applications and talent across the University.

    I am definitely willing to pull my weight in this endeavor. Where do we sign up? ;-P

  24. Jeff Carouth on April 29th, 2009
  25. This basic train of thought came up in another meeting today.

    One thing that we can do… instead of complaining that uweb is dead, let’s start hitting the listserv with discussions, questions, and ideas and revive it as a campus entity.

    One of the reasons it has languished is that most of the group’s leadership has left the university, so maybe it’s time we bring in some fresh faces and revive interest. I think it would be very effective to have an unofficial group of webmasters from across campus actively advocating things to balance what our group can do officially.

    One caution – we might want to get separate conversations, or even separate lists, going for application development and webmastering. They are both different beasts with different needs. Our departments run the spectrum from having professional IT staffs updating their sites to having student workers and secretaries doing it; we don’t want to scare off those who need our help most by keeping the conversation focused only on programming.

  26. Erick on April 29th, 2009
  27. Jae – interesting that you mention software licenses. We have a deal with the College of Architecture whereby we provide the license for the Atlassian suite of products ( and they provide the server hosting and administration. It has been a collaboration that has worked well for both of us.

  28. Erick on April 29th, 2009
  29. As far as licensing what what-not.. Vendors would absolutely hate it, and purchasing would find it a nightmare (consider fund sources and constituent budgeting), but wow it would save bones!

    Imagine.. a registry sniffer (not unlike a Microsoft/Novell audit) that unapologetically collect software use information university-wide and store in a central db for easily exposing cost savings opportunities.. Maybe a network sniffer for discovering common applications (like Jira/Confluence..)

    Here at the county, IT makes most IT-related purchases, and we thereby can centralize and bulk-purchase for the distributed enterprise. We are a much, much smaller org, but we still have state/federal grants that necessarily exclude some purchases from this potential.

  30. bronius on April 29th, 2009
  31. There are several good ideas being thrown around here, but we’re dancing around a bigger problem that exists at the University. Certainly, knowledge dissemination is always something that’s highly encouraged and highly useful. Software code sharing amongst departments typically isn’t a problem because there isn’t a bigger licensing monster we have to deal with. But as already mentioned, how do the departments continue to collaborate effectively to ensure either has the latest copy of changes they submitted? It is a challenge.

    I take a look at Universities around the US and find it interesting that we as technology enthusiasts always refer to Virginia Tech, University of Washington, UC Berkeley, etc. as examples of great technology innovators (particularly with software). Certainly they may have programs tailored to turn out software engineers, but it’s what they do on the peripheral that helps benefit the world – open source software. We all know these universities because of this fact because likely each of us are using some piece of software developed by at least one of them (or more). Texas A&M, however, isn’t in this pool. Sure, I can find some examples of software that’s open source (look for library & open source in the GSA), but that software is for a niche and not widely distributed. We as departments and colleges have similar problems that likely extend well-beyond the boundaries of the University. Furthermore, we’re not developing software to make money off it – we’re developing it to solve a problem. Why not release it as open source software?

    I’ve been considering taking a different approach than just simple knowledge dissemination – a Community Developer Network. In short, we should break away from the typical model of “we’re a 3 man developer shop” to being a part of a larger developer community, working on projects we all have needs for. This is certainly not without its problems. By know most people are thinking about whether their boss would ever consider the idea of allowing their employee to work on projects that wouldn’t perceivably benefit the college or department. There are plenty of arguments both negative and positive to this approach, especially at the beginning of something like this.

    Nevertheless, I believe considering the visions the Provost and the President have for the University (Academic Master Plan, Research Roadmap, and Vision 2020), this type of solution would actually help to put Texas A&M on the map as a community of serious innovators and ultimately help boost the University’s ranking amongst public institutions.

    Should anyone have thoughts or suggestions, please e-mail me offline (cweldon at We’re starting to put the gears on the wheel for something like this, and ultimately would like to find where the rest of the developer community sits on this kind of idea.

  32. Christopher Weldon on April 29th, 2009
  33. Chris, thank you for your post. I think you touched on a topic Robert and I briefly discussed earlier in the week regarding structure in terms of departments, colleges, groups, etc.. I completely agree that in order for the ideas presented here to be successful we must break out of roles as isolated development silos. As you hinted this will definitely not be an easy task, but I believe—given time—we can make the best tasting stone soup around.

    However, I do not know that we can easily take on the bigger problem that we are dancing around without a demonstrated interest in collaboration. I have already seen several cross-organization efforts, but they are few. I think that we should model our approach along the lines of the fable of the stone soup—otherwise out soup might actually be only water and stones.

  34. Jeff Carouth on April 29th, 2009
  35. Please, please, please take the Uweb wiki and listserv and do with them what you will! As Erick mentioned, several of the founders have moved on (or moved to Qatar, which makes it difficult for them to get interested in meetings). The rest of us have new responsibilities that have eaten up the time we used to devote to Uweb. I would love to see a fresh group infuse some energy into the project (or revamp it as something else entirely).

    I think Erick is also right about the different needs of application developers vs. web designers. Some of us cross over a bit; others don’t at all. I love Chris’s idea of a developer network. Just keep in mind that there needs to be something for the HTML/Javascript/Photoshop crowd as well.

    Jae, I completely agree about collective buying — and those are the same thoughts that brought Uweb together in the first place! I don’t see how to make it happen without a strong central IT office, though… and while CIS does a good job of handling software licenses for students, they just don’t offer what departments need.

  36. Stephanie Leary on April 29th, 2009
  37. I didn’t know that about UWeb. That sounds like a good place to start then. I also think the idea of keeping web page building and application building areas in different areas is a good thing.

    I agree with Jeff that I don’t think we can jump right in to the deep details without making sure we have something to build with first.

    I also didn’t know about that Atlassian license. That’s good info we could be sharing (even if it’s just that that’s what you’re running).

    My group is working on some IT-related websites for the fall, but for now, I think starting with UWeb is good.

  38. Craig Thompson on April 29th, 2009
  39. Back on the Atlassian licenses, Marcomm and the College of Architecture are more than eager to have more groups collaborate with us and jumping into our instance, as opposed to just buying the product themselves. There are already a bunch of different groups running their own instances, none of which are clustered. I think if we get a large number of development and IT teams in a single instance, we can keep scaling the infrastructure and ensure high availability.

    Let alone the fact that when one group buys a license for software, everyone benefits from it. 😀

    Ping myself, Erick Beck, and/or Ben Floyd if y’all are interested in getting into what we’ve got. That or simply seeing a demo of its capabilities.

  40. Christopher Weldon on April 29th, 2009
  41. Everyone, thanks for your posts, I’m glad the blog was able to serve as a catalyst to get things off the ground. And I’m glad to see the response we had to this article, it was exactly what I was hoping for when I posted it.

    I’d suggest that we now move this conversation over to the uWeb listserv so that we can continue the momentum and draw in those who might be subscribed there but who don’t see the blog and these comments.

    I think the next step ought to be to call for those who are interested in taking up the leadership torch for uWeb to get together in one room and start talking about how we can carry this effort into the future.

  42. Erick on April 29th, 2009
  43. Yeah, Qatar here 😉 Steph and I are owners on the UWEB listserv. If anyone wants to volunteer to be an owner, I’ll be glad to add you. Regarding Application Development vs. Webmastering, we can setup Listserv keywords on the UWEB list if you guys think that’d be helpful…

  44. Court on April 29th, 2009
  45. Erick, was the catalyst for you reading The Pragmatic Programmer?

  46. Robert Stackhouse on April 30th, 2009
  47. It was on Ben’s desk and after fighting with a broken water pipe in the yard all week I didn’t want to spend the weekend doing more of the same. It had a few nice stories, but to tell you the truth, after the first two chapters I actually wasn’t all that impressed.

  48. Erick on April 30th, 2009
  49. Erick,

    I suspect quite a bit of the information in that book is old hat to you. It would probably strain credibility to know the number of developers out there to whom that information is not old hat. That book, Mythical Man Month, and The Design of Everyday Things should be required reading for Computer Science students. I am appalled by tales of how much trouble some of my student worker colleagues have trying to convince their fellow students to use version control in Upper Level projects.

    Maybe Practices of an Agile Developer or Ship It would be of more use to you. You could ask Ben. I’m pretty sure he’s read both.

  50. Robert Stackhouse on April 30th, 2009
  51. […] is high time we mend our ways.  Thanks to Erick for giving me an adequate push in the form of this post to get me off my laurels and finish writing […]

  52. Overloading :: Robert Stackhouse on April 30th, 2009
  53. I’m definitely interested in helping take the leadership torch. One site my group (Communications and Marketing inside Texas A&M Information Technology) is planning to build right now is for IT pros, and we want to get the content from IT pros on campus, so this is a perfect opportunity at a perfect time. But this group, with all of the feedback so far, could be soooooo much bigger and better than that!

    I’m going to send a message to the listserv next.

  54. Craig Thompson on April 30th, 2009
  55. […] between people.  The best kind is face to face, but a persistent easily locatable record of a conversation that others can stumble upon isn’t a half bad substitute in some cases.  Perhaps we should […]

  56. What does it take to sustain a rich community? :: Robert Stackhouse on April 30th, 2009

Leave a comment