CEOP show Maddie is missing on 30th April 2007

"its def some kind of error" . . . rather leaves open the question.
Humans can 'commit' errors, by posting things on site before they were supposed to !
Computers do not 'commit' errors. They follow instructions given to them in their programming - created by humans
Dogs do not 'commit' errors. They may give an alert which humans, with their inferior senses and scientific equipment are unable to verify or explain. . .

He was also asked at one point whether he believed that Kate and Gerry were innocent of any involvement in Maddie's disappearance. And he said.....
....he would be astonished if anything came out to implicate them.

The exact quote is in this forum somewhere!

The way big Jim has fawned over the Mc's, endorsed that crock of a book, got in twitter spats, defend, protect and support from the get go at literally all costs which in itself is most bizarre tells us that either.....

a) He is incredibly naive and stupid...

or...

b) He is in it right up to his neck in all this (and he knows it).

My money is on b.

The connection between Gamble, the Mc's (Tapas) or some 'high profile' individual/s around the vicinity at that particular time could just crack this case.

IMO etc..

I still don't go with this pre-planned stuff that was orchestrated before the holiday commenced.

Something happened and Gamble was brought in to 'help'. By who and exactly when and how that 'conversation' happened is the £10 million plus question.

Find the links and find the answer.....

Perhaps.

Just my own theory on this red hot beautiful morning.

Just wondering if this little anomaly should go in the Portugal Resident ?

Mo wrote:Just wondering if this little anomaly should go in the Portugal Resident ?

They just might, although I think the PJ know already.

I've had time to look at the Dr Roberts search data a bit closer so got some more info.

2495 URLs are for 30/04/07 only (start and finish date)
520 URLs start with 30/04/07 but end at a later date (in a range)
Of the 2495 only 104 have the exact same identifier in their url as Madeleine 01 jpg & Madeleine 02 jpg therefore were captured at the same time.
Of the 104 only the Madeleine jpgs appear once the rest are replicated in other URLs (identifier is different but page name remains the same) the others are replicated 20 times on average.
.What does this all mean? Not 100% sure at the moment!

Skyrocket`s latest post over the way reproduces a letter, and reply, to/from :-

"University Professor (PhD) who is also president of his countries 'Association of Research into IT'. I don't feel it would be protocol to name him but his areas of expertise cover:


'His research interests cover the broad scope of digital libraries and information spaces, including specifically text and music information retrieval and organization, information visualization, as well as data analysis, neural computation and digital preservation.'

I don`t understand it but you teccies may do.

Was just reading that myself Mimi. Not that it made a great deal of sense to me.

I did notice (again) what a load of spelling mistakes were made for a Professor. Obviously just a coincidence


Here it is:

Dear ******

Your question is a very good one, but unfortunately not one that is straight-forward to answer, as it touches upon a lot of the technicalities related to Web crawling:
First, a webpage is usually not just a single file to be downloaded, but consists of several elements (text, inline images, ...) Depending on how the crawler downloads these it may happen that some parts are downloaded at a later date than the main text part of the page.
This issue has been discovered and resolved in later versions of the crawler used, prioritizing e.g. in-line images in the crawling queue, downloadi them immediately. So it will depend on the version of the crawler used.
Secondly, there have been many incidents of data having been lost due to disc failures (such a web archive is huge, so it is not easily feasible to keep several redundant copies). I have noticed in the past that when requesting a page at a given date that should be available, the server sometimes returned a different date (visible in the URL), serving most users' needs (and hiding the issue of data loss) but being at fault with good archival practices. I have critizized tis befire when meeting people from the IA, and I am not sure whether this is still the case or whether this re-direction has been made more explicit by now.
Thirdly, it is an issue of data management: I have too little insights into the internal operations of the IA, how time-stamps are assigned and managed, and the kind of errors that can happen at this level. I am pretty sure that the IA does not compute any cryptographic hash keys including timestamps over their data, so the concept of authenticity which usually is a cornerstone of a proper archive will be way harder to establish and verify than if you had the procedures that are in place at, say, The National Archives at the UK. (But then, it would probably be pretty hard and significantly more expensive to offer such services for something at the level of a global web archive)


Depending on what page is affected it might be possible to find it in another (possibly more properly managed) archive?
Also, if the question is merely whether a certain page existed as opposed to the content of the individual components it was made up of, then the situation is somewhat different: it would feel strange to find a webpage crawled with a certain date which should not have existed at that time stamp. This would hint at a much more complex error in the crawler set-up and should be reflected in zillions of pages. Also, this issue of mere existence might be discoverable from the source pages that contained a link to the page in question, i.e. the page that led the crawler to it in the first place, so existence is different from content.
For anythin beyond that, the log files might provide evidence of when a certain oage was actually crawled, but I am not sure whether the IA keeps those log files, and I have certainly never seen any of these made public.


Also, the IA has no globally valid mandate and legal basis for storing and redistributing webpages, so anybody can reauest the deletion of pages that they can provide evidence for of being the owner - again, a significant difference to a proper archive.


In a nutshell, it really depends on the structure of the page, the crawler used, and any erros that may have happened during data management, apart from the fact that any page can be deleted at the owner's request (if I remember IA's policies correctly - I did not have any more intnsive contact wih them for quite many years by now)


I hope this helps somewhat.


best regards,

********


I have since checked out the National Archives UK but although there are some archived ceop.gov.uk pages, none help (I don't think).

Hope this adds something to the mix.

Skyrocket1
Guest

Further to my post on the Dr Roberts link:-
The Madeleine jpgs (photos) are interesting as they follow the exact same process as the other jpgs in the index. There are individual crawls on single days and date range where a number of cross references are recorded. It appears that the other URLs are repeated by page name on numerous occasions (only on 30/04 however there are 2495 in total for that day). I think the Madeleine photos were captured on 30/04 it may be their file type that stopped the repetition (only my guess as they were the only jpgs captured that day)

Re Skyrocket's friend -
He's basically saying it's not a "proper" archive, because people are allowed to request the removal of things.
And bits of webpages can be crawled at different times, then put together to look like one web page, so what you are seeing is not always "accurate" given one specific day.
And that if the server breaks down and data is lost, then it gets "filled in" with whatever can be found.
He also seems to think it's better now than it used to be, though (he doesn't give a timescale).
He says he doesn't know for certain how the timestamps are generated, but that they aren't likely to be encrypted or hashed (that's expensive) as it's a free service and you can take it or leave it as regards authenticity of dates.
But he also says that it's very, very unlikely to find a timestamped page that should not have been there - in other words, the timestamp is not likely to be wrong.
To be sure, you would have to look at the server logs, which WB might or might not have kept. And which we're not likely to get.

Pretty much what's already been discussed earlier in the thread. I also think that English might not be the Professor's first language, so maybe we should cut him some slack.

Is there an easy way for me to upload a pdf file from a local drive, I'd like to demonstrate why the Madeleine 01 & 02 jpg are imo sound for captured on 30/04

WHO'S THE WE JIMBOY?

Poppy wrote:WHO'S THE WE JIMBOY?

What? Can you expand please

Hongkong Phooey wrote:Is there an easy way for me to upload a pdf file from a local drive, I'd like to demonstrate why the Madeleine 01 & 02 jpg are imo sound for captured on 30/04

Did you do this?




If so, how? Are you experimenting with Heritrix?

Hongkong Phooey wrote:

Poppy wrote:WHO'S THE WE JIMBOY?

What? Can you expand please

His tweet hkp,why did he say WE and nor I.

Not I Not nor I

-------

https://maddiemccannmystery.forumotion.co.uk/viewtopic.forum?t=823