In business and in international development circles, much is made about the potential for 'learning from best practice'. Considerations of the use of educational technologies offer no exception to this impulse. That said, 'best practice' in the education sector is often a rather elusive concept (at best! some informed observers would say it is actually dangerous ). The term 'good practice' may be more useful, for in many (if not most) cases and places, learning from and adapting 'good' practices is often much more practical -- and more likely to lead to success. Given that many initiatives seem immune to learning from either 'best' or even 'good' practice in other places or contexts, it may be most practical to recommend 'lots of practice', as there appears to be a natural learning curve that accompanies large scale adoption of ICTs in the education sector in many countries -- even if this means 'repeating the mistakes' of others.
But do we really need to repeat the mistakes of others? If adopting 'best practice' is fraught with difficulties, and 'good practice' often noted but ignored, perhaps it is useful instead to look at 'worst practice'. The good news is that, in the area of ICT use in education, there appears to be a good deal of agreement about what this is!
Here's a list of some of what I consider to be the preeminent 'worst practices' related to the large scale use of ICTs in education in developing countries, based on first hand observation over the past dozen or so years. I have omitted names (please feel free to fill them in yourself). The criterion I used for selection was simple: The given worst practice was easily observable in multiple prominent initiatives, with (one fears) a high likelihood of re-occurrence, in the same or other places. In no particular order:
1. Dump hardware in schools, hope for magic to happen
This is, in many cases, the classic example of worst practice in ICT use in education. Unfortunately, it shows no sign of disappearing soon, and is the precursor in many ways to the other worst practices on this list. "If we supply it they will learn": Maybe in some cases this is true, for a very small minority of exceptional students and teachers, but this simplistic approach is often at the root of failure of many educational technology initiatives.
2. Design for OECD learning environments, implement elsewhere
With the best of intentions, and often 'assisted' by vendors, many groups (including many governments) have sought to simply transfer ICT-related models and practices from classrooms in industrialized countries to less developed education systems in other parts of the world. Sometimes this works, but unfortunately many places roll out programs and products that have at their core sets of assumptions (reliable electricity and connectivity, well-trained teachers, sufficient available time-on-task, highly literate students, space to implement student-centric pedagogies, relevant content, a variety of cultural norms, etc.) that do not correspond with local realities. The result is often (and not unsurprisingly) not very good.
3. Think about educational content only after you have rolled out your hardware
Deploying lots of computer infrastructure in schools is expensive (and complicated). So expensive, in fact, that many critical complementary investments (in training, in tech support, in content, etc.) are 'postponed' until a later date. Sometimes this is a calculated bureaucratic maneuver/risk -- the thinking is that, once the hardware is in place, the need for content will be more clear, and it will be easier to make the case for related funding at that time) -- and other times this is simply a lack of good planning. But it is a fact that, in many places, only once computers are in place and a certain level of basic ICT literacy is imparted to teachers and students is the rather basic question asked: What are we going to do with all of this stuff?
Related to this ...
4. Assume you can just import content from somewhere else
Some places recognize the need for quality educational content from the start, but assume they can simply import it from somewhere else. In addition to obvious potential cultural issues, the successful integration of content developed elsewhere into daily teaching and learning practices is inhibited by a lack of clear understanding by teachers of the relevance of such materials to the required curricula. Much effort typically needs to be expended to map this content to explicit objectives and activities in the local curricula. (And of course: Teacher training helps too!)
5. Don't monitor, don't evaluate
This should be self-evident . That said, there are only a handful of really credible, rigorous impact evaluation studies done of educational technology initiatives in developing countries. Most evaluation work  focuses on (perceptions of) changes in attitudes as the result of the use of educational technologies, and the success (or lack of success) in meeting various simple metrics (number of computers installed, number of teachers trained, etc.). Such information is important, of course, but it is hardly sufficient. What is the impact of ICT use in education?  If we don't evaluate potential answers to this question, rigorously and credibly, all we are left with is well-intentioned guesswork and marketing dross.
6. Make a big bet on an unproven technology (especially one based on a closed/proprietary standard) or single vendor, don't plan for how to avoid 'lock-in
Let's acknowledge that the speed of technological changes almost always outpaces the ability of educational planners to keep up. In response, some policymakers seek to get 'ahead of the curve' by placing large bets on new, largely unproven technologies in an effort to 'leapfrog' what is happening in other education systems. In other cases, education systems effectively outsource most of the capacity to manage activities in this area to a vendor or other third party. There are potentially valid reasons to pursue such courses of action in some cases, but they are inherently very risky, especially if clear plans are not made on how to 'exit' such decisions and relationships.
7. Don't think about (or acknowledge) total cost of ownership/operation issues or calculations
What does ICT use in education cost?  Some people would have you believe it is only the initial cost of hardware. Businesses have long known that this is not the case, but many education policymakers seem not to have grasped (or willfully ignore) this fundamental issue. We know that "total cost of ownership or operation" (TCO) is often underestimated, sometimes grossly, when calculating costs of ICT in education initiatives in developing countries. Estimates of initial costs to purchase equipment to overall costs over time vary widely; typically they lie between 10-25% of total cost. That said, there is a dearth of reliable data, and useful tools, to help guide education decisionmakers in their assessments of the true costs of educational technology initiatives.
8. Assume away equity issues
One compelling justification for large-scale investments in the use of ICTs in education is that they can help address equity issues related to the 'digital divide '. That said, introduction of ICT in schools often exacerbate various entrenched inequities in education systems (urban-rural, rich-poor, boy-girl, linguistic and cultural divides, special needs students -- the list is long). Things can be done to mitigate such challenges, and indeed pro-equity approaches  of utilizing ICTs are possible, but they don't happen without careful proactive attention to this issue.
9. Don't train your teachers (nor your school headmasters, for that matter)
If there is one clear lesson from the introduction of educational technologies in schools around the world, it is that teacher training is critical to the success of such initiatives. Outreach to teachers, through both regular technical and pedagogical support and on-going professional development, should be seen as cornerstones of any large ICT investment in schools. And: Targeted outreach to school principals is often crucial if teachers are to have the necessary freedom to take advantage of new opportunities offered through the use of ICTs.
[I thought I would leave #10 blank as an acknowledgement that there are many additional worst practices that merit mention, but I have run out of space. Do feel free to submit your candidates below.]
For those who work in educational technology, none of these will be new. For many others new to this topic, the items on this list may appear to be so obvious that they need not even be mentioned. Even if indeed they are 'obvious', that unfortunately hasn't stopped them from occurring (and re-occurring) around the world with depressing regularity.
Other recent items related to ICT/education of potential interest from the World Bank:
1. infoDev has released a preview of the key findings  [pdf] from its upcoming Survey of ICT & Education in India & South Asia.
2. You can now RSVP to firstname.lastname@example.org for the following two events at the World Bank in DC in May (both start at 12:30pm and are open to the public):
- 5 May: Interactive Radio Instruction: Increasing Access to Quality Early Childhood Development Programs in Developing Countries 
- 13 May: Uruguay's Plan Ceibal: What happens when every student and teacher has their own free laptop