As with Executive Summary I think this section needs some work, as it will be one of the most read bits. Headers are a bit confusing (either we need table of contents, more consistent headings, or something to make this easier to understand what is going on!).
Generally, clearly a lot of work has gone into this report, and here is the bit where we present our findings - the major themes, areas that need attention, opportunities for progress - not just cobble together various points that the people we spoke to have made on lots of different topics. Hence this should be streamlined, edited, and thematically clear, with some kind of narrative to guide the reader through the various points.
Can we do a better job at linking the points we're making to the case studies we have? E.g. when we make a claim about something being an issue, can we point to case studies that mention this, and really spell out links between this section and other sections of the report?
I'd be inclined to remove two tiers of headings, and just have one subtitle size, and flesh out each subtitle to really spell out what our finding was (and how this relates to subject of report - CSOs and their use of spending data).
Comments on text:
"key takes" -> key points
"Demand for data" -> are we basically saying 'these are the types of data that we've found people are interested in'? also - as noted in another issue - do we want to mention revenue (which came up in a video interview)? In which case we might want to call this "What kinds of public financial data are civil society organisations interested in?", and say a bit about generally what different groups found most interesting, why and what the obstacles are for each kind of data?
Again thing about subtitles with different levels.
"(* See note)" convention is a bit weird. Can we either do this in a conventional footnote, or work into text?
typo: "information biographical information"
"Proactive spending release doesn’t necessarily make strong civil societies" -> I think this is a point that should be sufficiently self-evident to most people reading this (at least from the transparency side, but also from many on the open data side), that I wonder if we can think of a better way to reframe it so we're saying something more unexpected? Sometimes there are users without data, and sometimes there is data without users. This isn't necessarily just because of technical literacy issues, I think also it may be the case that for many groups spending data isn't sufficiently interesting to merit attention (in a world where this is in finite supply) - especially relative to other things, like advocacy around important issues, or looking at meatier documents/information sources, not to mention good old fashioned human sources. Perhaps our narrative can be something like 'in order to achieve accountability through information on public money you need the right kind of data (about issues that matter, sufficiently granular, etc) and data literate users who are able to make sense of it to do something' - and turn this into something about training needs and citizens and civil society organisations being able to coalesce around and use data effectively.
"Data supply" and "There are clear leaders amongst governments regarding publishing practices" - as this is a report about CSOs, I think we should frame this in terms of CSO engagement with governments, rather than as recommendations for governments (which it currently reads as).
"A few improvements on spending transparency occur for reasons other than public pressure" - perhaps change this to "There are a wide variety of factors that lead to the release of better spending data" or "Public pressure is only one of many reasons to the release of better spending data"
Can we combine "No bulk download" and "Governmental PDF release policies" into "Lack of machine readable, bulk downloadable data is an obstacle to transparency and accountability". Also would be good to mention specific examples of this from case studies above.
"Privacy Concerns" - again can we flag anything about this from the report above? Otherwise it just looks like this is something we think is important, rather than this being about CSO needs/use of fiscal data. Could we turn this into being about privacy vs transparency balance and mention Farm Subsidy legal case and how this is an emerging area of importance?
"Arbitrary release thresholds" - > perhaps we can turn this less into something about how these thresholds were arrived at, and more about engaging civil society around the transparency policy-making and commitments? We could allude to the OGP here.
"Don’t reinvent the wheel" -> perhaps mention how we need better mechanisms to exchange experience and expertise around civil society projects on public finance. I wouldn't make this about tech, but also about sharing knowledge (e.g. also on tactics, FOI, as well as engagement around data). Then we could rename to something like "Better exchange of expertise and experience between civil society groups working with information about public money".
"More community mapping projects!" - suggest integrating this into point above.
"Teach people" - can we flesh this out into something about civil society data literacy/data skills?
"several skills are repeatedly emerging" -> skills are lacking?
"While the pool of skills is hugely diversity across the community, key parts of the data pipeline from data retrieval to data cleaning are major hurdles, which steal time from activism and other parts of an organisation's work." -> bit confusing, perhaps too many different metaphors going on here (pool, pipeline, cleaning, hurdles)
"On the methodology and terminology" - I'd collapse this all into a section on how we need to join up CSO community (traditional transparency and more tech/data oriented groups/orgs) and turn this into something about collaboration, and shared understanding, not just about words like 'open' and 'standard' (which are really very minor points).
"So please if you are a polyglot, who can help bring an otherwise undiscovered project to the fore, put us in touch (info [at] openspending.org)." -> feel this is somewhat misplaced and we should take this out. Generally I don't think it is appropriate to ask for volunteers throughout text - as this should really be about CSO needs, not OpenSpending project needs!
To finally conclude are there any specific recommendations we can make (e.g. to funders, to transparency orgs, to data/tech/civic hacker groups?), or next steps we can suggest? Would be nice to distil what we have learned from this report right at the end, so others know what we think needs to happen next if we want to maximise impact of CSOs using spending/public finance data...