We welcome any query on Who When Where. If you have previously posted it on another forum (including the old WDYTYA forum), please state this in your opening post - this will save people redoing the research which has been done before: they can look at it and possibly go further with it.

Online GRO Indexes Queries Raised

A space for genealogy-related conversations.
Mick Loney
Posts: 371
Joined: 15 Jun 2020, 07:27

Re: Online GRO Indexes Queries Raised

Post by Mick Loney »

The real shortfall with the GRO site, is its the lack of options when searching.
I found a birth on Freebmd, but it only showed an initial. To find the second forename, I went to GRO site. I put the same details into the search, and nothing came up, obviously a mis-transcription somewhere. What was more frustrating, was that despite having the full reference, I couldn’t make use of it, as that is not an option GRO allows.
You can omit all details when searching, except surname, so if that has been mis-transscribed, your stymied!
User avatar
AdrianBruce
Posts: 358
Joined: 14 Jun 2020, 18:57
Location: South Cheshire

Re: Online GRO Indexes Queries Raised

Post by AdrianBruce »

Hardwork wrote: 25 Mar 2021, 17:15 ... I think it clearly shows the set up has been created by people who have very little understanding of its function and who refuse to acknowledge the weaknesses inherent it.
I think that's quite right. The restriction to +/-2y, and compulsory completion of M/F are stupid restrictions. Indeed, from an IT viewpoint I can't imagine any justification whatsoever for such tight restrictions. Anybody would think it's running on a 386 PC. I can't imagine either what use cases went into the requirements analysis (i.e. what did they think we were going to do with the system?). I suspect that as far as they are concerned, genealogists are not the intended customers but people who want certs for themselves or their close family and therefore know roughly when they were born.

However, we need to be realistic - what is the cost benefit analysis for breaking down thing like the +/-2y and compulsory completion of M/F? I'm not sure. While the cost is surely minimal, the benefit to the government is zero. Reverting to the ability to double index certificates, conversely, is almost certainly a major job. There's one surname index on each entry - generating another, according to the previous rules, rewriting the SQL (or whatever) - probably a big job.
Adrian Bruce
User avatar
AdrianBruce
Posts: 358
Joined: 14 Jun 2020, 18:57
Location: South Cheshire

Re: Online GRO Indexes Queries Raised

Post by AdrianBruce »

Mick Loney wrote: 25 Mar 2021, 19:14 ... You can omit all details when searching, except surname, so if that has been mis-transscribed, your stymied!
Yes, which comes back to the question of what the user requirements were - and I suspect, as I say, that genealogists were not in the user requirements. Not defending it, just stating my belief.
Adrian Bruce
AntonyM
Posts: 66
Joined: 15 Jun 2020, 12:44
Location: Buckinghamshire, UK
Contact:

Re: Online GRO Indexes Queries Raised

Post by AntonyM »

I'm not going to defend GRO and their index, but I can perhaps add some explanation. I was one of quite a number of genealogists and researchers that were invited to consultation meetings with GRO when the new index and the provision of PDFs were introduced in 2016.

All indexes have issues and faults ( and the new index, like the old, is known to have missing entries for various reasons) - and all have different rules about how material should be indexed. The "new" indexing of the birth/death records was done many years before the information was used to populate the on-line GRO index. The indexing was done through a 3rd party outsourced arrangement as part of the ill fated DOVE project in 2005. It was done by going back to the (copy) registers held by GRO, and compiling a fresh index, not by just digitising the existing indexes.

The rules used to create the written/printed indexes changed over time - with the later addition of age to death indexes, and the inclusion of mother's maiden names on birth indexes in 1911. Likewise the rules given to the transcribers involved in the DOVE project were not the same as those used for the old printed indexes.

The major difference does seem to be around the indexing of births to an unmarried couple, but where both parents are named. In the old indexes these are indexed under both surnames, but in the new index, if a father is named on the entry, then that is the only surname the entry will be indexed by. This seems to be the case in the example at the start of this thread - and also explains why GRO don't feel any amendment is necessary - the indexing is correct under the rules being used. I understand they get a large number of "errors" being wrongly reported just on the basis that the results don't match FreeBMD, which is to be expected given the changes, and is why they won't enter into correspondence on them. The only errors that really get acted on are where they check the entry and find there has been a clear mis-transcription of a name.

Another difference in the on-line GRO index is that births to unmarried women now have their maiden names (correctly) shown as blank, rather than copying the odd decision made in 1911 to simply repeat the woman's surname in the maiden name column. Another is the inclusion of middle names in full, rather than just initials (as on some of the printed indexes).

At the consultation meetings I did ask to see a copy of the indexing rules that were used for the transcription, but GRO were unable to supply, or even find them (apparently). My suspicion is they were probably drafted by a business consultant working on the outsource project for Siemens who didn't fully understand the nuances of registration, but I have no evidence for that. What consultation, if any, took place with the research community at that time (2005) I have no idea.

Regarding the search restrictions - these were subject of a lot of debate and protest at the consultation meetings. In response GRO were very keen to stress that the "new" index was designed as an ordering aid, and not as a research tool, but most importantly they said they were very concerned about the potential impact that wide/vague searches may have on the performance of their existing systems ( there was no funding available to upgrade them at the time in advance of such an anticipated large increase in demand). They did promise however, to review the restrictions once the impact had ben evaluated - whether that happened or not, I don't know, but the restrictions remain in place.

Personally, I find the differences between the old and new indexes very useful - once you understand the differences it means that by comparing the results you get on FreeBMD with the result on GRO you can can use that to work out the whether a father is probably named on the entry, and the marital status of the parents, something which is impossible (with 100% certainty) from a single index on its own.
Last edited by AntonyM on 26 Mar 2021, 12:09, edited 4 times in total.
Professional Researcher (retired) and former Deputy Registrar
AntonyM
Posts: 66
Joined: 15 Jun 2020, 12:44
Location: Buckinghamshire, UK
Contact:

Re: Online GRO Indexes Queries Raised

Post by AntonyM »

A few points worth remembering when searching birth indexes:

1) Before 1969, there is no surname shown for the child. The only surnames shown are those of the parent(s), and as discussed the entries are indexed by one or both of those surnames (depending on marital status and which index you are using).

2) Marital status of the parents is made clear by looking at the informant column ( especially after 1874). If both parents sign together as joint informants, then you know they are not married. If a single parent signs (father or mother) and the father is named, then you know they are married, or at least claiming to be.

3) The appearance of maiden name in the indexes doesn't always mean the mother is married to the father, or even that a father is named at all on the entry.
Professional Researcher (retired) and former Deputy Registrar
User avatar
AdrianBruce
Posts: 358
Joined: 14 Jun 2020, 18:57
Location: South Cheshire

Re: Online GRO Indexes Queries Raised

Post by AdrianBruce »

Antony - thanks for your comments. At least it appears that there was some consultation with genealogists, etc.
My suspicion is [the indexing rules] were probably drafted by a business consultant working on the outsource project for Siemens who didn't fully understand the nuances of registration, but I have no evidence for that.
Hmm. That matches my guess that the GRO may not have realised that they'd changed the rules.
GRO were very keen to stress that the "new" index was designed as an ordering aid, and not as a research tool, but most importantly they said they were very concerned about the potential impact that wide/vague searches may have on the performance
That's an understandable concern - in fact, it would be a dereliction of duty not to be concerned. However, there are multiple ways of dealing with that, and I suspect that the customer (the GRO) designed the technical solution (tight restrictions) then presented that to the contractor, who imagined that these were business requirements and - who knows - designed a system that actually can't cope with bulk searches - instead of designing a system that (e.g.) might cope with bulk searches but might also say, "Sorry - too many". Put simply - Civil Servants should not design IT systems - that's what they employ people for.
Adrian Bruce
Post Reply