Wikipedia:WikiProject Military history/Peer review/List of castles in England

List of castles in England
This list has been thoroughly revamped in recent months. I should like to know how the list now measures up to reviewers' expectations of a list of castles in England. Paravane (talk) 16:37, 6 July 2011 (UTC)

MarcusBritish

 * Although the use of a Key for English Heritage and National Trust looks good, I think it would be more useful to reorganise the Ownership column with Alpha-codes, for example:
 * EH - English Heritage
 * NT - National Trust
 * PR - Private Residence
 * PP - Private with Public Access
 * PV - Private No Access
 * LA - Local Authority
 * HA - Holiday Accommodation
 * HT - Hotel
 * CE - Church of England

Then make it sortable. Reason I suggest this is, many people like to visit castles, so being able to sort by ownership and find English Heritage and National Trust sites, or whatever type interest someone, easier could be quite useful.
 * Has this approach been adopted in other similar cases, are there even standard codes? Using just codes would be neat, but some points:
 * It would increase dependence on the key for understanding, and there are more possibilities to consider - for instance 'Private with Public Access' would need to be split to distinguish free access to private land versus opening for restricted hours. It could be confusing and would be less readable because of the number of codes and the need to refer to the key.
 * - I don't think there are standard codes, but as Wiki articles are generally "self-contained" I don't think it will be frowned upon if you were to invent them, as long as it's understandable. I was thinking the code column should only contain the code - everything else should be moved into Notes regarding which councils, residents, limited hours, etc. If necessary you could make it a 3 letter code, PRO - Private Access Open, PRL - Private Access Limited - however I would advise it might be better to keep it simple and simply identify access restrictions in the notes. Castles could change hands (new residents, new trust/authority ownership) so it is always good to design a table that is manageable - whilst you might be happy to keep an eye on it and update anything you learn, you never really known if/when your wiki support might wane and require someone else to edit the contents without "breaking" the layout due to confusion. Ma &reg;&copy; usBr iti sh '''  (talk)


 * There's freedom for variation currently, with named owners and combinations of ownership and stewardship. You could have for instance 'LA - Somerset County Council', or 'CE - Residence of Bishop of Carlisle', but is this an appealing format?  Alternatively, some extra information could be transferred to the notes column.
 * - Yes, I think if you move all information other than the code into Notes, so it becomes a sorting column and key, rather than anything else it would work well. Ma &reg;&copy; usBr iti sh '''  (talk)


 * It could be made sortable without using codes, the images could be combined with abbreviations EH, NT etc. without opting for other codes, but sorting would not group local authorities etc. Paravane (talk) 21:19, 7 July 2011 (UTC)
 * - As long as all Councils are marked "LA", for example, then they would group during sorting - not sure what column it would sort by following that though. But as you already have the tables split into Counties then as long as LA's group they should be manageable. Ma &reg;&copy; usBr iti sh '''  (talk)
 * - Additional: One way of making LA's group by their respective council names alphabetically would be to have something like  in each LA cell, and then include details of the Council visibly in the Notes column. I can guarantee this method works great as I have a similar use of in an article I created recently. Same could be done for Private Residences, only put the owners name like    in the Notes.. and so on and so forth, wherever there is need to control how it sorts Ownership an extra step beyond just the alpha-code. That way, the sorting result will always appear logical, and this extra step might also push the article an extra step towards A/FL quality a bit earlier. As for the National Trust/English Heritage logos — I suppose they would become redundant if you switch to an alpha-code system, but you could still use them in the Notes or straight below the code for extra visual identification. All your NTE_icon.png's need switching to NTE_icon.svg's if you do keep them — the PNG versions have white backgrounds, oddly, but the SVG versions are transparent, which looks much better on the coloured table background.  Ma &reg;&copy; usBr iti sh '''  (talk)

The current ownership info largely predates my involvement, and I'm now reminded that there's a lot of information still missing.
 * - Yes, seems like a big job what with nearly 300 entries on there to manage. I might also suggest adding a new column: "Grid Ref" and using the  template (Results in  for Windsor Castle, for example, to identify the location of each castle geographically on OS maps. That's a lot to do also (adding 1 per minute is ~5 hours work) but I think it would be a very strong addition to the material because it is practical modern information for tourists, etc looking for these places rather than just historical data. And it's not "bulky" content - i.e. only 8 chars and the external logo.  Ma &reg;&copy; usBr iti sh '''  (talk)


 * There is a very long References list, perhaps using  or   to make it shorter and page-width would be better for the layout.


 * Perhaps setting the column widths of each table to eliminate the current "stepping" effect would improve the visual layout. Or simply make each table 100% width and let the columns sort themselves so that they are at least left and right aligned.
 * Not sure if this can be improved. On my machine at 1024x768 all tables are full width.
 * - I have 1366x768, a laptop screen, and believe me, there is a lot of stepping on the right at this width. Ma &reg;&copy; usBr iti sh '''  (talk)

Points:
 * Forcing a narrower table to 100% seems to leave the image column wider than the images - even if that column is set to 90px - which looks very poor.
 * - You may need to account for whatever cell-padding or margins wikitable's have and use 100px or so. Let me know if you want me to take a look, as I've used a few wikitables and with the benefit of wider screen might be able to get it to fit okay, as I can down-size my screen resolution to see the result on 1024x768 also. Ma &reg;&copy; usBr iti sh '''  (talk)


 * If the sizes of columns are fixed, I'm not confident that this will not have adverse consequences for smaller format devices such as tablets.
 * - I don't think it should as long as it's done carefully. Ma &reg;&copy; usBr iti sh '''  (talk)


 * Maybe the content could be tweaked to improve layout! Paravane (talk) 21:19, 7 July 2011 (UTC)
 * - Well, bear in mind, if you use a 2 or 3 letter code for Ownership and move all other info into Notes that you will reduce that columns width quite a lot and gain some room for the rest. The extra info in the Notes will not widen that column either, but will heighten in line with the image column so it will balance out. On a side note, I don't think the line break in each column heading to push the "sort" icon below the heading is a particularly good look - unless it benefits it by minimising the col width, that is. Personal choice, though. Ma &reg;&copy; usBr iti sh '''  (talk)


 * Hyphens need changing to emdashes (WP:EMDASH) in: "Once no longer needed as fortresses, castles - if they were not abandoned - were..."


 * Similarly, all date ranges, such as "11-14th century" or "1892-1930", need endashes, "11–14th / 1892–1930" (WP:ENDASH), not hyphens.


 * I feel a few more citations may be necessary to cover a few points, and perhaps a bit more detail. There is a lot of text, but the 9 in-line citations are a bit thin, imo. Few examples:
 * History
 * "During the fourteenth century, as much of England became more peaceful, the construction of strong castles began to decline, in favour of more lightly fortified structures often described as fortified manor houses." - Why did they decline? Evidence of change in style?
 * "In the reign of Henry VIII, prompted by fears of invasion, a series of new fortresses was built along the south coast of England, known as the Device Forts or Henrician Castles." - Invasion from who, and why? Evidence of series being constructed for this purpose?


 * I have R.Allen Brown's book on Castles too, only mine is: - not sure if you have a US version or if it's a different title altogether. If you need any extra sourcing from it, feel free to ask.

Lovely article, though - lots of potential for A-class, FL ratings. I think it still needs a little work in places to make it easier to view (layout tweaks) and some further work to the historical references. Stronger sourcing and a lot more citations in the text should push it along.

Ma &reg;&copy; usBr iti sh ''' (talk) 22:51, 6 July 2011 (UTC)


 * I've done a couple of these suggestions for you to save time — there were almost 300 ndash replacements, to save you wading through them I've done them all in one go with a regex, plus a couple of other minor tweaks — no content changes, just MoS copy-edits. Ma &reg;&copy; usBr iti sh '''  (talk) 00:11, 7 July 2011 (UTC)
 * Thanks for your detailed comments and edits. I'll need some time to address citations.  I have responded above to other suggestions. Paravane (talk) 21:19, 7 July 2011 (UTC)
 * - Thanks. Replied to some of your comments, above. Ma &reg;&copy; usBr iti sh '''  (talk) 23:47, 7 July 2011 (UTC)

I've had a look at your current set-up, here's what I think would work well for the layout of each table - just my 2c though, do what you think best, depending what level of quality you're aiming for at present. I'm suggesting FL quality tweaks here, where everything counts.:

What I've done in this example:
 * Fixed column widths - they do not exceed 700px with the Notes col left to fill the remained of the "100%" fixed table width. Incidentally, I had no problems when testing the Image column at 90px fixed width.
 * 90% font-size, it's barely noticeable, but by compressing the content just a little, it becomes more manageable for wider tables.
 * Examples Alpha-code and OS Grid Ref (I've just noticed List of castles in Scotland uses OS Refs)) columns. Example of EH logo under Alpha-code to add visual recognition.
 * In notes I've put anything relating to Ownership as a bullet point - not sure if this works, but it emphasises ownership a little more.

In addition - the current Dates column does not work as a sortable column at present because it contains dates, ranges, circa years and centuries all mixed up. Although the information is displayed as necessary to read visually, it does not sort logically. It should be possible to come up with a method of sorting the Dates logically using the "display:none" method again. Just to clarify what I mean though:

Finally, for all the circa dates, the WP:MOSABBR requires use of "c." for circa, rather just a "c" before the date being estimated. There is a Template:Circa which I'm not recommending you bother using, as it does very little, but it's style uses "c. " with a space before the date which seems right to me, grammatically. I'll go ahead and do this in a moment.

Hope this all helps! :) Ma &reg;&copy; usBr iti sh '''  (talk) 05:08, 8 July 2011 (UTC)
 * I appreciate the time and thought you're putting in to this.


 * Grid references have always been under consideration, but for me were low priority: most sites have their own wiki pages with coordinates already, and pages for the rest are gradually being created. Adding an extra column will have an effect on layout at lower display sizes and as there are ~450 entries it will be a big job.  You say it would be a strong addition, but do you not think anyone interested will access the wiki page for the site and get the location from that page?
 * - Adding empty columns/templates is easy - a regex can do that it seconds. Filling in the Grip Refs would take the longest time, a few hours. at least. Yes, I think it's worth it - OS Grip Refs link to Geographical Data sites with mapping, satellite images, etc - all that might prove useful to people - it's "encylopedic" data. The layout should not be dramatically affected - cutting the Ownership column down from its current width to a narrower width for just 2 or 3 letters would provide the space needed. If you look at this article: List of Sites of Special Scientific Interest in North Yorkshire you'll see there is a Grid Ref col - there are dozens of those SSSIs articles with hundreds of sites listed, many of which are just patches of grass - and probably far less interesting than many castles. It's the sense of "completion" that such data provides, and that it strengthens the article by providing practical data. Anyone can read about a castle, but being able to pin-point it, perhaps to visit, is a bonus. If you've got all the castles in England on one page in front of you with OS Refs, it saves you having to search a dozen other sites for the information. Ma &reg;&copy; usBr iti sh '''  (talk)


 * Sortable dates is plainly another task to be added to the todo list, your example works well.
 * - Those can all be done with a couple of regex's in less time too, would save you doing ~450 manually. Ma &reg;&copy; usBr iti sh '''  (talk)


 * Table widths - I'm glad you supplied an example, because I do not think it works very well at all! At 800x600 it's not bad, but at 1024x768 the columns to the left are large and mostly empty whilst the notes column is almost the smallest, full of squashed text forced to wrap to multiple lines, and causing the rows to be deeper than the images.  At 1280x1024 the image column is wider than the images again.
 * I have tinkered with the content of the tables to try to reduce the stepping, but I cannot go higher than 1280x1024.
 * - I noticed. The use of  between words only serves to stop line-breaks and maintain wider cells. This has the negative effect of preventing cells from fitting narrower screens, because the words will not fall onto new lines. On wider screens such as 1366x768 the stepping is still evident however. It's not just the right edge - the columns are not equally aligned as you scroll down, making it a little awkward to read - visually it's not too appealing in the way it dots about. 100% fitting with fixed widths is usually the only way to control tables, on any website. Unfortunately there are a lot of screen resolutions and it's not always possible to fit them all perfectly. I can understand wiki editors wanting their own contributions to look good on their own screens, but it is always necessary to consider a wider range of common screen sizes - with flat screen monitors and laptops starting to outnumber PCs, there are more wider screens, 1024x768 is the most common CRT monitor resolution. One way of making the notes less "stacked" might be to make them   size notes and not too specific - afterall, if each is cited and wikilinked, there's no need to put more that a few key points in the Notes to highlight important points. If the Image col is fixed width, it's not possible for it to go wider at higher resolutions. Another thought might be to make the Names/Type/Condition columns a little narrower and let them fill 2 lines, to give more room to the Notes column. Either way you do it, there is a lack of consistency between one table and the next which I think needs to be addressed for the article to achieve a high standard, to look professional it needs to flow neatly - which is going to require fixed widths, and some sacrifices or alterations to compensate for the amount of data each table holds. I have 1366x768 which seems wide to me, yet I know there are far wider resolutions out there, becoming quite common - their view will be even more "stepped" than mine - the only way to eliminate that is 100%.  Ma &reg;&copy; usBr iti sh '''  (talk)


 * My view is that the single thing which does most to enhance the appearance of the tables is to keep the images fully occupying their cells.
 * - I agree a "full" cell looks better than one with spacing. Again, it won't always work on every resolution, no matter what you do. And when you weigh the "value" of data in each cell you have to determine what is more important - a picture or sufficient textual data. In the end a few spaced images won't "break" its appearance. There are going to be visitors out there with eyesight issue who make their browser text larger, and that will affect every cell anyway. There's no point in worrying too much about vertical width really, there's virtually infinite amount of that, it's horizontal width that needs controlling. Ma &reg;&copy; usBr iti sh '''  (talk)


 * Circa - A small point, but in your layout example you have c.1458, no space. I think this looks much better than c. 1458, does this conform to the approved format?
 * "c. 1066" is right, there should be a space if Template:Circa is anything to go by. I don't think anyone would cry about it if you omitted the space, though - it still means the same thing. Ma &reg;&copy; usBr iti sh '''  (talk)


 * Ownership - the dates approach could be applied to the ownership column, the codes could be the hidden text. This would enable full sortability whilst retaining the current visible text and images.
 * - Wouldn't work as well. Just a code would sort by that code, then by another, probably Name resulting in a logical sort order. If you had hidden codes then text it would sort by the code and then by first letter/s of the visible Ownership text, making it no more efficient than it is now. Ma &reg;&copy; usBr iti sh '''  (talk)

I favour this option:
 * It's more readable, without the need to memorise the key or keep referring to it, or the possibility of misunderstanding.
 * - Current Ownership is a wider column and has no standardised method. A code would make it more logical. People wouldn't have problems with a key, really - people adapt, or sort and search for things they specifically want. Ma &reg;&copy; usBr iti sh '''  (talk)


 * It's flexible.
 * - For who? Readers of editors? Ma &reg;&copy; usBr iti sh '''  (talk)


 * It avoids mixing ownership information with other information in the notes column
 * - I don't see this as a negative point, personally. Ownership: Private. Side-note: Owner is Sir John Smith. Seems standard to me and more logical use of the Notes column which is unsortable, allowing better sorting of the Ownership - primarily to extract Accessible castles, such as those run by EH, NT and LAs. Ma &reg;&copy; usBr iti sh '''  (talk)


 * Using just codes makes sparse use of the column, given the width required by the name.
 * - Putting the Sort icon below the "Ownership" heading would make it narrower, or renaming it to "Owner" would help. It's a trivial matter compared to the benefits of a logical Sort key. Ma &reg;&copy; usBr iti sh '''  (talk)

Not sure what is normal for table headers and sort buttons, I have no particular preference. Paravane (talk) 22:22, 8 July 2011 (UTC)

Just another example, of possible display methods, from playing about with tables and sizes. This one combines Name/Image, also to see how it might add more space vertically as well as heighten each row. Again, just throwing rough ideas down; food for thought. Ma &reg;&copy; usBr iti sh ''' (talk) 07:26, 9 July 2011 (UTC)

I've added the Date sorting info so that all the dates, centuries, etc sort logically like in my earlier example - I don't think it's broken anything whilst doing that. Whilst manually correcting a couple of things I noticed that there is a lot of use of Template:convert for tower/wall heights, etc. Only problem here is it's very inconsistently used - sometimes it's going from ft->m, other times from m->ft. I think one way or the other needs to be selected and applied throughout, so that it's uniform which appears first and which in brackets. I don't think there is a MOS standard as to which is best practice. Also noticed some heights use "ft" some "feet", again needs to be one or the other to neaten things up - especially for FLC standards - "ft" is probably best. Ma &reg;&copy; usBr iti sh ''' (talk) 08:48, 9 July 2011 (UTC)
 * Thanks again for your input. I think that's about everything dealt with except the two remaining issues of ownership and grid references.  A counter-example to the List of Sites of Special Scientific Interest in North Yorkshire is the List of tallest buildings and structures in London. Paravane (talk) 23:46, 10 July 2011 (UTC)
 * Not necessarily. People tend to use street maps and A–Z guides to find their way around places like London where directions tend to be more specific. Castles, often located in the middle of the country, which doesn't usually have A–Z coverage, are easier to pinpoint by a 6-digit Grid Reference, which would probably be less appropriate on articles like List of tallest buildings and structures in London. Ruins hidden behind trees or just as earthwork mounds to the untrained eye are unlikely to have a postcode whereas anywhere can be Grid Ref mapped and found on http://www.getamap.ordnancesurveyleisure.co.uk/ - which is the same for SSSIs, many are isolated patches in the country, not on public routes, hence Grid Refs being valuable. Like you said yourself, accessibility is sometimes questionable for many castles - OS maps tend to highlight private land/roads, as well as public access routes and NT/EH sites - speaking from experience, as I enjoy hill walking myself and have visited a couple of castles this year as there are plenty in the north, near me. Including virtually all of York. Ma &reg;&copy; usBr iti sh '''  (talk) 01:11, 11 July 2011 (UTC)
 * The concept behind this list has not been to provide a rigorously comprehensive listing, as there are more than 2000 sites - see for instance The Gatehouse. A complete listing of all sites where there are clearly visible remains is broadly the aim, but only those with 'significant' remains are listed in the main tables, creating a browsable collection of all the most interesting sites - which can be elusive when distributed amongst vanished castles and earthwork remains.  This leaves open the possibility of pursuing completeness in separate county listings, already done for Cheshire, for example.  For all the earthwork sites which are not in the main tables, there is no provision for location data in the chosen format, but this is available in the county lists, and is also available for every site which has its own wiki page.  So if you want to visit Sigston Castle, say, this list as currently conceived cannot supply the location data.  The ownership column is now sortable, albeit with a different logic to your original proposal. Paravane (talk) 23:10, 12 July 2011 (UTC)


 * Indeed, I cannot imagine anyone wanting to tackle an article with 2000+ sites. This article can sometimes be a little slow to load already, even on broadband, I think many more entries would make it impossible to load given the high number of images and tables it has to render. It is probably best to keep the list discriminatingly selective.
 * I think the article might be almost set for a BCR once you reference a few more of the details in the text. I think it is a small step to B-class, but I expect it is a long way from A-class and and FLA. I, personally, do not follow the logic of the Ownership column - I find myself referring to the key more often because the icons are not easy to remember and having a mix of icons and text seems rather inconsistent. I don't think this will be an issue for BCR - but for anything higher, I expect it might lack support. My main concern, however, is that you have applied a logic that might not be immediately apparent to other editors or inexperienced contributors. Hence why I think a simpler key would have been much more appropriate. It would be easier for any other editor to pop in "PR" for a Private Residence, and detail the owners name in the Notes, than have to check if there is need for an icon or not, and the way that particular column is formatted. At least with a consistent format there is less room for error. For any chance of FLA, and possibly ACR, I expect a major overhaul would be required to support both readability and simpler editing.
 * I disagree with your doubt about OS locations. Earthworks are listed on OS maps, sometimes the word Motte in Gothic-script appears on a map when referring to an old long gone castle where nothing remains but humps and bumps in the ground, or a few old remnants of wall a few feet high. Plenty of "Roman Ruins" are labelled in the Lake District on my map too, in much the same way. I passed through one of these "mottes", once. If it weren't for the note on the map, I wouldn't have recognised it as anything more than a deep ditch in a sheep field. But more to the point, there is only one format for OS Grid refs, a 4–8 digit number, depending how accurate you like to be. Most people use a 6-digit grid ref. There's no reason why every single castle on the list cannot be found on a map. I am unsure why the editors of the Cheshire castle list have chosen to use the coord template - that seems all well and good for people using Google Earth or satellite data online - but your average walker, tourist, rambler, etc doesn't use latitude/longitude - they use a Grid Ref - which leaves this article open to go one better than that and provide practical data for outdoor use, not online use - which makes far more sense. Nevertheless, that article is FLA, due to its consistent layout, and inclusion of data such as geolocation - this article has the same potential, imo.
 * So far it's a fine article, well developed, with room for more development. Seems to have been a lack of peer reviewers, beyond myself, which seems a shame - I would recommend you stick in a few more citations to strengthen the referencing side of things and then proceed to BCR and see if that gets a few more ideas from reviewers, beyond what I have already suggested would be worth doing extra.
 * Ma &reg;&copy; usBr iti sh ''' (talk) 04:23, 15 July 2011 (UTC)
 * As long as you do have icons, I recommend  = &dagger; for those that are Church of England owned castles. It's identifiable.  Ma &reg;&copy; usBr iti sh '''  (talk) 09:52, 15 July 2011 (UTC)
 * The article has already been assessed as B-class. Regarding the ownership column, your confusion with the icons demonstrates how something that seems logical to one person may not be so easily comprehensible to another, which - thinking of readers rather than editors -  might be equally true of the two-letter codes.  Apart from a few unusual cases, if a castle has an icon it is regularly or frequently open to the public, otherwise it is not; the icons themselves are a 'standard' resource, used in other articles beside this one.  Logic aside, arguably the icons work at an intuitive level, standing out from the text and drawing the eye to the castles open to the public, rather than the rest.  I did not myself initiate the use of icons, but I have now extended the logic of it.
 * I do not doubt the usefulness of grid references, the issue with the earthwork castles is that in the case of those which are simply listed above the main tables in each county, there is no place for a grid reference as the list is currently conceived, unless each name is followed by a grid reference in brackets, say, which would be unsightly.
 * Thanks for all your thoughts, edits and suggestions, I feel that some reflection and perhaps a wider input of views would be sensible before any more substantial changes are made. The most useful work currently on castles in England is perhaps in creating new wiki pages for all those castles on the list that do not yet have one.  Paravane (talk) 20:12, 15 July 2011 (UTC)
 * I didn't notice that it was already rated. I don't think it meets B-class, personally, I don't think the referencing is strong enough for an article this size. Not the tables - but the text preceding them has too many uncited claims for my liking. I think it needs re-evaluating to allow the new material to be reviewed - can't expect any article to keep the same rating if it has been heavily rewritten. I note the update was performed by Woody here quite recently, but I don't know how he considers all major points as being referenced when that is clearly not the case; there are at least a half dozen things should be referenced - those being uncited historical claims, most of the current citations relate to construction/architecture/state of ruins, etc. Bearing in mind that this is a WP:MILHIST peer review, I think it important to point that out. If you need me to tag those things as cn let me know, but I think it's pretty obvious what needs sourcing, so I see no need to uglify the article.
 * I wasn't thinking of readers - I was thinking of both readers and editors equally - I don't think the Ownership is easy to follow and even less so in terms of updating. 2-letter codes are no more than abbreviations - something used in many things - postcodes, reg plates, initials, etc - people have a natural knack for relating letters to words. I still think the sorting order is not as strong as it could be either, simply because mixed icons and text don't allow anyone to comprehend the order. Whilst a logo of English Heritage might well be sorted as "EH", without that distinction, it isn't clear. I have an affinity for logic, mainly from years of web coding, as well as using OS maps that do use icons to identify things - but I still find some elements of the tables less straight-forward than they could be.
 * Just my 2c though. Ma &reg;&copy; usBr iti sh '''  (talk) 07:31, 16 July 2011 (UTC)
 * Regarding OS grid references: I'm the editor who put together list of castles in Cheshire. coord was chosen as when used in conjunction with GeoGroupTemplate it allowed the reader to click through to a map off all the castles in either Bing or Google maps. I appreciate OS grid references would be useful for UK readers, but those not from the UK might not have a clue what it means. That said, I imagine the majority of the audience will be from the UK so I'm not too fussed either way. Perhaps both types of co-ordinate could be used? As for sourcing, a couple of additional references might be welcome, but the content itself is correct (I know WP:V is policy and I'm not saying more references shouldn't be added, but that the hard work has already been done and that adding a couple of inline citations shouldn't be too hard). However, being familiar with the subject matter I'm not seeing which points need further references so it might help if you list the instances here rather than adding cn tags to the article. Nev1 (talk) 16:54, 18 July 2011 (UTC)
 * I don't know if both would be practical, as the OS references also link to online satellite data - I think Google make Grid Refs translate into longitude-latitude coords, but not vice versa. I don't think anyone not from the UK will actually be noting coords, so the use of Grid Refs wouldn't make any difference. Not that I can picture many people wanting to see a castle from space, they're the type of thing you want to see from ground level, personally, not a vague outline from above. But I don't think it's unfair to say, why would Wiki have created a Grid Ref tag that only makes sense to Brits, if they don't expect us to use it? I personally can't make sense of the way longitude-latitude coords are written, it is more complicated - but if both types link to the same data, my preference would be to utilise OS Refs for the bonus that they can be used by walkers with regular OS maps. Ma &reg;&copy; usBr iti sh '''  (talk) 03:51, 19 July 2011 (UTC)

Citations
 * "Some were later rebuilt in stone, but there are a great many castle sites in England where all that is visible today are traces of earthworks."
 * - citation to support that castles were rebuilt.
 * - citation to support visible traces exist today.
 * Done. Nev1 (talk) 13:09, 20 July 2011 (UTC)


 * "In the far north of England, where conditions remained unsettled, fortified buildings continued to be built as late as the 16th century, not only by the rich and powerful but by any with adequate means, as defence not against great armies, but against the notorious Border Reivers."
 * - citation to support that hostilities in the north existed, hench the requirement for more castles.
 * - citation to support their need to defend against Border Reivers, wikilink alone is not adequate.


 * "A castle differed from earlier fortifications in being in general a private fortified residence, typically of a feudal lord, providing the owner with a secure base from which to control his lands, as well as a symbol of wealth and power."
 * - citation to support this difference between earlier defence, and show of wealth and power.
 * This is already referenced to Brown 1962, it's just that the reference is attached to the following sentence. Nev1 (talk) 13:09, 20 July 2011 (UTC)


 * "The many Roman forts of which ruins survive in Britain differed in being wholly military in nature, camps or strongholds of the Roman army: the Romans also built town or city walls in England which can still be seen, for instance at Silchester."
 * - citation/s to support the Roman influence on fortification designs, given that this is the only key mention of the Romans in the article, something material to support their role in this area of English history.


 * "By the 16th century the role of fortifications had changed once more with the development of artillery capable of breaching even thick stone walls."
 * - citation to support development of artillery, to prompt the need for greater castle strength; this is an important evolution in castle design, due to the early use of gunpowder and cannons, and should be strongly referenced.


 * "Once fortifications had become altogether redundant, it became increasingly rare in England for new buildings to be described as castles,"
 * - citation to support the reduced use of the term/description "castle".


 * "—were, over the centuries, adapted and modernised to make them more suitable for continued use as residences:"
 * - citation to support that developments were done in past centuries, rather than by modern organisations to preserve the appearance.


 * "In the 18th and 19th centuries especially, many castles underwent 'improvements' by architects such as Anthony Salvin, and in this period a fashion developed for entirely new houses to be built in the style of castles, and to be known as castles"
 * - citation to support that Anthony Salvin existed and did this work, wikilink alone is insufficient.
 * - citation to support that such a fashion developed, as this is a historical claim, someone must have written about it.


 * "No list of castles in England can ever be complete, because there will never be complete agreement in every case as to whether the remains of a building are those of a castle, whether a given place is the site of a castle, or whether a surviving building should be considered to be a castle."
 * - citation to support this claim; who says there can't be an agreement? Obviously there are experts in the field of castles, cite one who can support this.
 * It seems like common sense as border issues are always a problem, but I have added a source. I would also note that new excavations can uncover new sites, as was the case in the 1940s for instance (if you wish to include that, it can be referenced to page xiii of Castellarium Anglicanum). Nev1 (talk) 13:09, 20 July 2011 (UTC)

Regards,  Ma &reg;&copy; usBr iti sh [talk] 07:04, 20 July 2011 (UTC)


 * Paravane, do you want to take care of these or can I chip in? I've only skimmed over them, but I reckon I could take care of the bulk in half an hour tops if my heap of castle books isn't too disorganised. Nev1 (talk) 10:27, 20 July 2011 (UTC)
 * I'm very happy for you to help take care of this, I have limited resources on my bookshelves. Thanks!  Paravane (talk) 12:27, 20 July 2011 (UTC)
 * I'm getting error messages from the Wikimedia Foundation when I try some of my edits so I'll return to this later. Nev1 (talk) 13:09, 20 July 2011 (UTC)
 * I'm sure you're both doing a great job; sorry to be a pain, but the higher the citation standard, the better the article - Paravane cited a few points I raised earlier in this PR (seems like a long time ago, doesn't it? Wiki suddenly becomes a FT job..) so the handful I raised here were all that really stood out to me as major points that need sourcing. If I can just make a minor observation, though. You added 2 new books and gave them ISBN-10 codes: in WP:ISBN they prefer the use of ISBN-13 (even to the point of stressing it in bold) where possible. If the book is quite old, and doesn't have a 13-digit ISBN at the back or on the Copyright page, I normally find it's easy to get any off Amazon.co.uk just from matching the title/author/year/format, as it's their Primary key. If not found there, simply Google "ISBN converter" - top result is a tool than converts ISBN-10 to -13 and vice versa; you can also enter ISBNs in the Amazon Search box to confirm them, it doesn't have to be a just title or author you search for. NB: ISBN format simply needs to be "978-1234567890" with no hypens other than the one after 978. Just mentioning this, as it helps maintain consistency with the other titles referenced in an article, is up to date with the latest ISBN standard, and helps get articles promoted quicker by following citation standards (gives quality reviewers one thing less to complain about) - I'm forever doing it as a routine part of copy-edits, especially when converting older articles book lists to the newer cite book template. Cheers,  Ma &reg;&copy; usBr iti sh [talk] 14:55, 20 July 2011 (UTC)

Hchc2009
Really good to see this page developing! Some thoughts from me:
 * There are some long paragraphs early in the article, such as:
 * "Pele towers which became known as castles, or preserve a castle-like aspect, are included in the present list. Many others, or their remains, have survived much altered, incorporated in later country houses or farmhouses, and are excluded from the list: amongst these are Aske Hall, Biddlestone Chapel, Bolling Hall, Bolton Old Hall, Boltongate Rectory, Causey Park House, Clennell Hall, Cliburn Hall, Corbridge Low Hall, Cowmire Hall, Craster Arms (Beadnell), Croglin Old Pele, Denton Hall, Dovenby Hall, Dunstan Hall, East Shaftoe Hall, Godmond Hall, Great Salkeld Rectory, Hardrigg Hall, Hepscott Hall, Hetton Hall, Hollin Hall, Hutton Hall (Penrith), Irton Hall, Johnby Hall, Killington Hall, Kirkoswald College, Levens Hall, Little Harle Tower, Littlehoughton Hall, Nether Hall, Netherby Hall, Ormside Hall, Pockerley Pele, Preston Patrick Hall, Randalholme Hall, Rock Hall, Rudchester Hall, Sella Park, Selside Hall, Skelsmergh Hall, Smardale Hall, Thistlewood Farmhouse, Warnell Hall, Weetwood Hall and Witton Tower."


 * I found these quite lists hard to work through, as we've effectively got lists turned into paragraphs. I could think of two alternatives:
 * You could put the list into a footnote. This would make the intro read more easily, but still include the information within the article.
 * You could create list pages for them (e.g. "List of castles incorporated into later houses") and wikilink to the list


 * As an aside, I knew what you meant by "how castle-like is the surviving building", but there might be a better way to phrase "castle-like"!
 * Hchc2009 (talk) 17:11, 15 July 2011 (UTC)
 * Thanks for your thoughts, it's good to have feedback from multiple sources! I'm sure others will agree with you in this case, but I have felt the approach is necessary in view of the inevitable arbitrariness in determining what sites should appear on the list.
 * I concede that I have created 'paragraph lists', which is an unusual approach. The reason I have created them is with a view to stabilising the main list as well as providing a context for it.  I would argue that for this list these are important functions which justify an unusual approach, function ultimately more important than style, and putting the lists in paragraphs keeps the information very compact.
 * These paragraph lists are assigned to a section entitled "Scope and exclusions", which warns the reader, if not to expect quite such long lists, at least to realise what is being done. There is no need to read this section at all unless 'scope and exclusions' is a compelling topic.  You say you found these lists hard to work through, but I wonder why you would choose to work through them: unless you have a particular purpose, surely you would just skip to the end of the list.  Given that each list is all in blue, it's easy to see where the end is.
 * None of these lists is separate from the list of castles itself, since they are all mutually exclusive, necessarily so. Once you create separate lists on separate pages, it will be impossible to prevent occurrences of the same building on more than one list, and their function in establishing boundaries and exclusions will be lost.  Putting the lists into footnotes would - I would argue - make them less prominent, less likely to be noticed, likely to be thought less important, and less likely to be effective.  By being taken out of the main text, they will effectively be demoted.
 * 'How castle-like is the surviving building' - there might well be a better way of phrasing it, but it's succinct, avoids trying to answer the question as to what this might mean, and is not invented! (cf. castlelike) Paravane (talk) 21:24, 15 July 2011 (UTC)

Nev1
According to D. J. Cathcart King's Castellarium Anglicanum, the most significant comprehensive index of castles in England and Wales, gives over 1,500 sites in England. While I mention that figure, I note it's reference to Liddiard (2003), but he edited the volume and the chapter being referenced was written by Richard Eales, so the reference needs tweaking slightly. In no way would it be realistic for a single list on Wikipedia to include all of those sites in a manageable form; just take a look at the Gatehouse website to see how long those lists can get. For this reason, I like the approach taken here to concentrate on those buildings with some surviving remains. I think it's a good idea to measure lists against the Featured List criteria, and this issue relates to criterion 3a: "It comprehensively covers the defined scope, providing at least all of the major items and, where practical, a complete set of items; where appropriate, it has annotations that provide useful and appropriate information about the items."

For the most part I think the article satisfies that, but I can think of at least one case where an important castle is relegated because it was utterly destroyed. Bedford Castle was the subject of one of medieval England's best documented sieges, and in the aftermath it was razed to the ground so as there are no remains (apart from a much mutilated motte IIRC) it's relegated to a note at the beginning of its section despite being one of the more important castles in the county. It's possible there may be other cases like this, but that one really stuck out for me. How to address that, I'm not quite sure. Maybe you should just note in the lead that there are going to be some exceptions to the rules laid down in the lead because whichever way you slice the cake you'll probably find some important sites outside your criteria.

On the subject of criteria, the article casts the net wide, which is a must really for this kind of exercise, but not too wide. I'm reminded of the preface to Adrian Pettifer's English Castles: a guide by counties:

"English castles are a very diverse group, and consequently it is hard to do them justice. They range from massive edifices which still dominate the landscape to 'motte-and-bailey' earthworks and Border pele tower. Castellated mansions of the later Middle Ages are often dismissed as castles altogether because of the concessions made to comfort, but castles always display a compromise between the requirements of defence and domestic convenience. I have adopted a more tolerant approach and accepted all sites with enough defensive characteristics. ... My aim is to give a brief description and history of every masonry castle except the most fragmentary. I have been more selective with earthwork castle because they are so numerous. ... Real castles are a medieval phenomenon, spanning the centuries from the Norman Conquest to the Tudor era, and that sets the limits for this book. Buildings and earthworks of other periods fall outside its scope, even if they are popularly called 'castle' like many ancient hillforts, Roman forts and stately homes. I have included the small proportion of medieval fortifications which are not actually castles, these being walled towns, defended monasteries and Tudor coastal forts. They are too closely related and it would be a pity to ignore them. For the sake of clarity a few sites which do not strictly merit inclusion have been allowed in, namely sham castles on genuine sites and medieval manor houses which bear the name castle but do not actually deserve the title."

Had I made this list, I would have stuck to King's list in Castellarium Anglicanum, and if Hchc2009 had written it probably would have taken on a different form from this or how I would have planned it. But there's more than one way to go about this, and I certainly appreciate this approach.

The history section seems fine. It provides a brief history of the castle in England without overwhelming the reader, but there are a handful of minor issues. Where it says "Castles continued to be built in England for several hundred years, reaching their peak of power and sophistication in the late 13th century" the bits in italics could be changed to "military sophistication", it's up to you. Where it says "Earlier fortified structures, such as the Saxon Burh or the Iron Age Hill fort, provided public or communal defences,[9] as did town or city walls which were built in medieval times" I'd ditch "which were built in medieval times" as it implies it was restricted to that period which you then contradict a couple of sentences later when talking about the Romans building town walls. Also shouldn't burh be lower case? The third paragraph interrupts the chronological flow; could this be avoided by making it the first paragraph and adding to it the first sentence of what is currently the first paragraph?

I think some of the lists from the scope and exclusions section can be trimmed. In the table section I notice a couple of maps; I'm not sure how practical it will be to have a series of maps. Some will become very crowded, and when included may cramp an already fairly wide table. For the most part the tables are fine, although I think a key would be useful. For example does date refer to when each site was founded, when building work took place, when it was in use, or some other significant event? And it needs to be made clear that ownership refers to current ownership. I'm not completely sold on the type column. With so much variety in castles it's not always easy to classify them, and as people try to be more precise in their classification more and more obscure terminology is developed, and there is also the issue that the form of a castle can change over time. What may have started as a timber motte and bailey may later find its motte levelled for example. But, as a quick reference I don't suppose it does too much harm. There are a couple of terms that appear in the table that aren't explained elsewhere, for example "shell keep" is dropped on the reader without explanation.

Sorry if I've repeated anything the other reviewers have said, I've only had time to read through the article rather than other people's comments. I am proud to have seen this article develop and I cannot commend Paravane enough for the effort that has gone into this. Nev1 (talk) 16:42, 18 July 2011 (UTC)


 * Thanks for your comments, answering your points in the order in which they appear:
 * The Liddiard reference was a stopgap, it would surely make sense for the reference to come from King, if he makes mention somewhere of how many sites he has included, but I have not had convenient access to a copy.


 * Bedford Castle - it could be given its own row in the main tables, I've done that for Queenborough. The focus has been on surviving buildings, to the neglect of military history.  If Bedford, which others - Bristol, Gloucester?  It would be preferable for the number of vanished castles in the main tables to be kept to a minimum, I think.  From the lead: "Vanished castles or those whose remains are barely visible are not listed, although exceptions have been made for some important or well-known buildings and sites."


 * Criteria / post-medieval sites - this list is included in WP:ARCH as well as WP:MILHIST, and of course Salvin and others worked on medieval buildings as well as wholly new ones in similar styles. I feel that including some sites that are not 'true' castles makes the list more comprehensive for general reference and usefully so.


 * History - when I wrote this, there seemed to be a reason for the order of the paragraphs, but I cannot now see any reason for ignoring chronology. Regarding "town or city walls which were built in medieval times", I originally had "the town or city walls which..." but 'the' was edited out by MarcusBritish.  Strictly, there is no implication that they were only built in medieval times, so no contradiction.  The train of thought was that hill fort and burh were comparable, amongst medieval fortifications, not to castles but to town walls; then it is a further observation that the Romans built them too.  I'll give a little thought to what changes might be made.
 * If I altered the actual meaning of any sentences with any of my minor edits, do not hesitate to correct them. My main intent was to make the text less verbose. Whilst I have no doubt about your good use of English, words like "moreover", "nevertheless" and similar words, often added padding rather than meaning to sentences. I cut a lot of them out, to make the text more direct and to the point, and also a little easier for non-English readers to handle, because these flowery words may not make sense to them. Ma &reg;&copy; usBr iti sh '''  (talk) 03:39, 19 July 2011 (UTC)
 * On reflection, the reason for departing from chronology is that this piece was intended as a quick and easy introduction to castles, rather than a history of fortifications. Originally it was the start of the article, the current lead was added later.  The motivation was to involve the reader directly with castles, and not distract or deter him/her by introducing hillforts or burhs or castra at the start - so more people might continue reading.  By delaying, the reader has been exposed to castles for a bit before being pushed to think about comparisons with other types of site.  Paravane (talk) 16:33, 19 July 2011 (UTC)


 * Scope and exclusions - not sure what you have in mind to be 'trimmed'! As I explained to Hchc2009 above, I see the lists in that section as having quite an important role.


 * Maps - these were only added yesterday, not by me, I have no immediate plans to contribute to them. I have put in a request for the template to be upgraded to offer a direct link to geohack from each location on the maps, which would make them more useful.  However, Northumberland and others would need to be rather large.  I don't know whether putting a full set of maps on a separate page would be an acceptable option.  A decision needs to be taken as to whether they are to be kept on the existing page.


 * Tables - explanations for date, ownership etc can be added to the key, above or below the icons. I really think current ownership could be assumed, but it's certainly better to be explicit. As regards type, I'm not sure whether you have reservations about including the column, or just the way I have done it.  There are some clear distinctions to be made, so I think the column should be there.  Keeping the information brief makes it easier to pick out essential differences between sites and helps keep the tables compact even at 1024x768, which is the resolution I work at.  This is just a reference list, with links to the specific pages which have more detailed information, my view is that rigour is not the only consideration in this context, though the descriptions certainly should not be misleading.  Compare describing the Earth as a sphere, it's not accurate but it's a useful starting point.  I was aware that 'shell keep' was not explained, I was inclined to the view that the term 'keep' has been introduced, explaining 'shell keep' will entail going into a greater level of detail, and any article has to make some assumptions about the level of comprehension of its readers.


 * Thanks for your support. I've spent a lot more time on this than I originally envisaged! Paravane (talk) 23:00, 18 July 2011 (UTC)

Dank
These are my edits; feel free to revert. I got down to Scope and exclusions. I don't have an answer to the question of how much, if at all, I should be attempting to copyedit for an international readership, and I'm open to talking about it. If you're not on board with these changes, please let me know.
 * FWIW, MOS asks for double quotes rather than single quotes, even in BritEng articles. ('castle' -> "castle" etc.)
 * Done.
 * "A few castles are known to have been built in England before the Normans invaded in 1066, a great many were built in the years following, the principal mechanism by means of which the Normans were able to consolidate their control over the country.": The first comma is called a "comma splice" on this side of the pond and I think over there too, but whereas they're still forbidden here, they're getting more common over there. I'd still recommend conversion to a semicolon.
 * Done
 * "A castle differed from earlier fortifications in being in general": I recommend "Unlike earlier fortifications, castles were in general ..."
 * I agree it's ugly, I'll give it some thought.
 * Changed wording.
 * "prompted by fears of invasion": it's not clear who or what was prompted.
 * Changed wording.
 * "a series of new fortresses was built": Both "was" and "were" will sound off to some readers; perhaps try "new fortresses were built in series", or just "new fortresses were built".
 * As it stands it is grammatically correct, both of your suggestions do alter the meaning.
 * No problem. - Dank (push to talk)
 * Changed wording.
 * "since they were not private residences, but national fortifications, they no longer possess": ... they did not possess
 * I would argue that the present tense is preferable here, in which case a future editor might make the reverse request. It is the present-day perception of these buildings that is at issue, those considered to be castles possess certain characteristics, in particular that they were built as private fortified residences.
 * "... a symbol of its possession; a castle becoming a grand residence proclaiming the status of its owner." comma here.
 * I originally had a comma, a previous editor changed it to a semicolon...
 * Changed back to a comma.
 * "... built by Salvin; a building so authentic ...": colon. - Dank (push to talk) 21:37, 1 August 2011 (UTC)
 * Done. I originally had a comma, a previous editor changed it to a semicolon, I'm not convinced that a colon is preferable to a comma...
 * Thanks for your interest and comments, I'll need to think about what changes to make. One thing that is clear is that successive minor copy-edits can generate slight but undesirable changes in meaning, and a process of creep.  Regarding your changing the criteria to a list of questions, originally some of them could be read as questions, but the context and lack of question marks indicated that they were not actually posed as questions.  I somehow think that this change might foster more changes, I can envisage that initially another editor may come along and object to the phrase 'list include such factors as' being followed by a list of questions.  Even though a list makes it easier to read, I am inclined to revert it, changing the word order, e.g. "how strongly fortified was the building" to "how strongly fortified the building was" - it wasn't intended to be read as a question, I just thought that word order was preferable. Paravane (talk) 23:32, 1 August 2011 (UTC)
 * Sounds good, and sounds like you're on top of everything. - Dank (push to talk) 00:55, 2 August 2011 (UTC)
 * Various changes made, there's the outstanding issue of tense referred to above, otherwise I think I have dealt with all your suggestions. Paravane (talk) 20:01, 3 August 2011 (UTC)
 * In peer reviews, I generally just do some copyediting and make a few suggestions, and don't follow up. - Dank (push to talk) 20:09, 3 August 2011 (UTC)