I happen to be the member of a few non-profit technology e-mail lists. One of them is the NOSI list. Just my luck, there was a discussion started by Dirk Slater who was asking about what kind of web cms should be used for the Community Voices Heard website. Dirk basically liked Joomla and Drupal as starting points. Little did I know that the NOSI list was populated by Plone vendors and developers. They really, really wanted Dirk to do the site in Plone.
And for what it’s worth, my org’s website runs on a Plone/Zope platform. We haven’t had the greatest luck with it but it works and is stable. However, I wouldn’t recommend using Plone/Zope for a smaller non-profit simply because it’s more of an application framework and not “just” a CMS. That little slash between Plone/Zope is ultimately my main concern about the platform in regards to a smaller non-profit. Zope is an application server and Plone is the CMS that runs on top of it. The application stack gets to be kinda complex to manage and administer when you’ve got Apache, Zope and Plone.
A quick look at Community Voices Heard’s Form 990 at Guidestar told me that they probably didn’t have the infrastructure to handle a Plone/Zope deployment: that is, there’s nobody in-house who will be able to watch over a complex Plone/Zope deployment and work with it exclusively. I’m not disparaging the Plone/Zope consultants but one does have to question whether or not it’s a good fit for an org with only $690,000 in expenses each year.
Eventually, I made this remark:
Basically, I think developers create a cycle of dependency when they implement a web app framework such as Zope/Plone for smaller non-profits. This is what I mean when I discuss the high IT priesthood (and as a former web consultant in the bubble era, I’m just as guilty of this as anyone). Many non-profits are then stuck in having to call the developer over and over again for minor changes such as a CSS style sheet change or a new content area with different content headings.
There’s no way that a smaller non-profit without dedicated IT staff can reach a level of robust independence from a developer if what they get is a web app framework and not an easy to use CMS.
And thus, my real fear. I’m not really interested in doing a critique of Plone so much as I am in the criteria that consultants and developers use in providing assistance to non-profits. As part of that discussion, I said:
As a general aside, we need to keep in mind the following thing as a IT worker in the non-profit worker [sic - I meant sector]: Are you sure you’re not creating future work for yourselves when you spec out a heavy-duty app framework like Zope/Plone for your clients and/or users? If they keep having to come back to you, are you really helping the non-profit sector? I think that if a client has to keep returning back to the consultant, that consultant has done a particular disservice. After all, you’ve just taken dollars that would have been spent feeding the homeless, helping people with life-threatening diseases or the myriad other things that non-profits too [sic - I meant to say do] .
Now, I’m not saying that all NPO tech consultants are opportunists looking to bleed NPOs dry. In fact, most open source tech consultants probably see themselves as the technical equivalent of NPOs themselves with the NPO standing in as the needy client. They try mighty hard to deliver solutions they see as the ethical equivalent of a food program.
However, there’s a life cycle that has to be considered when you’re building out what will eventually be an application. Traditional software development paradigms are fairly relevant when discussing the lifecycle of a website release. Once you start thinking that way, you then start thinking about the TCO involved with hiring consultants to support a product that is technically complex. Let’s face it, while the number is growing, it’s still more difficult to find developers and administrators for open source applications especially on Linux platforms. That tends to tilt the salary requirements for the in-house staff needed to take care of a Web site in the wrong direction. Combined with the low numbers of other capable consultants who work on the same platform, this will further constrain the options available to any smaller NPO.
NPOs have long ago seen this kind of interaction between themselves and their clients and have tried to alter this by changing the very nature of the interaction by doing more empirical analyses of the effects of their work on the outcomes of their clients and by reassessing the end product of their work and how it should affect their clients. I think it’s time for open source advocates to do the same.
I thought I was in the minority when I voiced my opinion (I even got flamed for even daring to speak poorly of Plone) but it turns out others share my opinions on this. A very reasonable position on OSS for NPOs is held by Nick Gleason from Citysoft. I think he and I are probably going to be virtually tarred and feathered but hey, this is what blogging is for. 
I really am looking forward to anyone’s opinions out there about OSS, especially if you’re a non-profit IT director/admin/worker. Have you experienced OSS’ peculiar brand of vendor lock-in? Does my critique apply to your ASP as well? If you want to keep from being eTarred-and-Feathered by leaving a comment, send me e-mail at confessions[type your at sign here]nonprofittechblog.org.
Loading ...
First, full disclosure… IT director for 5 years at the Eastmont Computing Center in Oakland, CA and now one of “those” consultants.
I have my own thoughts about Nick’s “very reasonable” position that he titles “the truth” about open source. I respect Nick greatly, but was disappointed to see them propagating FUD– fear, uncertainty and doubt.
But strip away the FUD and sales rhetoric and they have some solid points.
The real point that both he and you are making (IMHO) is that open source software requires strong technical skills. Technical skills cost money and if you don’t have the technical skills you can’t ensure your vendor doesn’t cross the line from off-the-shelf software to custom-off-the-shelf software to custom software.
So how does lock in work? If you buy Kintera and then want to convert to Get Active, you need to spend a lot of money on data conversion. And the vendors don’t help out by creating tools to assist with conversion, since they lose money when people convert.
In open source similar but different. If you want to convert from Plone to Drupal same data conversion problem, but the open source communities tend to build tools to support data conversion.
If the NPO doesn’t have strong technical capacity, the previous two paragraphs have absolutely no meaning. Nor should they.
How about if I implement CitySoft and decide that Nick and his crew aren’t meeting my needs? Among which vendors can I choose? None. I have the source code so I can maintain it myself.
Again, without strong technology capacity, the above paragraph is meaningless.
So why is open source going to make a difference? Because it is increasingly (not yet, but increasingly) becoming relevant to nonprofits without strong technology capacity.
Imagine purchasing a hosted eCRM system tailored to NPO needs. Click on an icon in the hosted solution and get a backup of your database (perhaps you didn’t like the custom service or your needs have changed). Grab a CD and install the same hosted software on your computer/server. Follow a simple wizard and have your hosted service and data running locally in minutes. Or do the same thing with another hosted service provider.
We can argue about what vendor lock in is all day long. That is definately not vendor lock in.
So when is OSS relevant to an average nonprofit? I would say when it is implemented as off-the-shelf or custom-off-the-shelf software. But if you are a nonprofit without strong technology capacity, you are just going to have to trust your vendor.
Great reply – thanks for taking the time to answer… Yes, I think I’ll better define what vendor lock-in means to me. I agree with you in that a hosted eCRM system is not true vendor lock-in. Fundamentally, vendor lock-in to me is a combination of the lack of transparency (technology being presented as divine secrets), a marketplace or developer community that is low in numbers so options for switching out a vendor is limited, and a lack of understanding sometimes by the client, sometimes by the vendor that the application cannot be maintained if the status quo were to change.
I don’t think a customized application is necessarily a locked-in application. It doesn’t have to be at all. I agree with you that when clients aren’t technically strong they don’t have the capability to see the difference between off-the-shelf OSS and custom OSS software. This is the line that I think many “heroic” developers and vendors cross when they ask clients to adopt.
I certainly don’t want to create FUD or make people think that OSS isn’t a great idea. I’m a big fan of OSS – in my spare time, I run this blog on WordPress and my friend’s blog at http://www.muckraked.com. I also have my own development website which is run on a Linux server. However, I’m very hesitant to recommend it to your typical NPO unless they have a very good record of finding and retaining good IT staff.
I’m the IT Director (and entire IT department) of a $2 million nonprofit, I just found this blog a few days ago, and I love it so far. Nice job.
RE: open source, what I like to do when looking for solutions is to totally ignore whether something is OSS or not. 95% of the time I’m not going to (or really shouldn’t) be messing around w/ the source. So, once that’s out of the way, I can compare features and TCO and see what best meets my needs.
Two examples:
–When choosing firewalls/security devices for our network, I chose generic PC hardware along w/ open-source OS and software. None of the proprietary appliances met my needs until the costs were in the many 1000′s of $, and here I get something well-tested, rock-solid, and w/ all the features I could ask for, from OSS. There was a learning curve, but it wasn’t bad (I’m a *NIX guy), and now it just works.
–When choosing email server software, I went the other way and chose a commercial package. Email is something that really needs to just work–it has to be up, and I don’t have time to mess around w/ sendmail (or postfix or…) config files. The package has a reputation for being solid, and support is included so if it blows up I get someone on the line to help me.
Now, I’m an OSS fan as much as anyone–we use it extensively internally, and I run some personal projects almost exclusively on OSS–but for the needs of our org, I find it helpful to look past it and compare solutions.
I need some info. When a company is operating on a 503c status, can someone with recommending authority purchase software from a company they hold interest in. This software could have a dependency which would create future purchases and upgrades for years to come. Is this a concern?
I need some info. When a company is operating on a 503c status, can someone with recommending authority purchase software from a company they hold interest in. This software could have a dependency which would create future purchases and upgrades for years to come. Is this a concern?
Sorry about resending. I did not click the followup by email.
Interesting question? I’m not a lawyer nor do I give out legal advice. However, as a matter of ethics, this issue is already covered in the SANS code of ethics at:
http://www.sans.org/resources/ethics.php
“I will avoid and be alert to any circumstances or actions that might lead to conflicts of interest or the perception of conflicts of interest. If such circumstance occurs, I will notify my employer or business partners.”
Just notify your boss.