Re: Claim by 'Stevo' - "CEOP show Maddie is missing on 30th April 2007"

by Joss 20.06.15 15:08

Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.

That makes sense.

by MRNOODLES 20.06.15 15:09

jeanmonroe wrote:"This error is proving quite a hard task for us to explain at the moment"
------------------------------------------------------
OG.

"This 'abduction', of Madeleine, is proving quite a hard task for us to explain at the moment."

My money's on, after they've 'fixed it'. It'll lead to another question that some eagle eyed person will ask after spotting another anomaly.

by Joss 20.06.15 15:11

jeanmonroe wrote:
Nuala wrote:But Nuala has already said it's impossible so it must be impossible because Nuala knows exactly how the Way BAck machine program worked in 2007 and that's why he/she signed on urgently to tell us.

Still no real explantation from the real people.

Right, that's it, I'm gone.

I'm not putting up with such rudeness when all I'm doing is trying to help.

Just chill, Nuala.

You know how many people have been 'rude' to me over the years?

NONE!

They wouldn't 'dare'!

I'm just getting to grips with 'Webscrumping', mentioned earlier.

Does that mean FREE 'virtual' apples?

by Joss 20.06.15 15:17

Nuala wrote:But Nuala has already said it's impossible so it must be impossible because Nuala knows exactly how the Way BAck machine program worked in 2007 and that's why he/she signed on urgently to tell us.

Still no real explantation from the real people.

Right, that's it, I'm gone.

I'm not putting up with such rudeness when all I'm doing is trying to help.

It's ok. Nuala, i have enjoyed reading your posts & btw, welcome2

by SixMillionQuid 20.06.15 15:19

Just chiming in briefly.

1. If this was a 'glitch' why would an automated system pick a precise date time of 30/04/2007 11:58:03? It makes no sense to make up a date before the the web page became accessible to a webcrawler.

2. If this was a 'glitch' then other pages archived around the same time would show similar dates and times. Wayback should be able to confirm this.

by Joss 20.06.15 15:25

SixMillionQuid wrote:Just chiming in briefly.

1. If this was a 'glitch' why would an automated system pick a precise date time of 30/04/2007 11:58:03? It makes no sense to make up a date before the the web page became accessible to a webcrawler.

2. If this was a 'glitch' then other pages archived around the same time would show similar dates and times. Webback should be able to confirm this.

That would be good.

by Syn 20.06.15 15:31

Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source code would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.

/**
* Construct and initialize structure from a 14-digit String timestamp. If
* the argument is too short, or specifies an invalid timestamp, cleanup
* will be attempted to create the earliest legal timestamp given the input.
* @param dateStr from which to set date
*/
public Timestamp(final String dateStr) {
setDate(dateStrToDate(dateStr));

full source here:
[You must be registered and logged in to see this link.]

by jeanmonroe 20.06.15 15:37

Oh well, to add to my list of 'things' i 'don't get or understand', like, 'Chinese Arithmatic' and 'where do all those bullets, that people fire in the air, in 'celebration', above their heads, 'go', i can now add, 'WaybackMachine.'

Went to a restaurant and asked 'do you do take-aways'?

the guy said 'yes, of course'

'one minus one........nuffink'

by **Richard D. Hall** 20.06.15 15:48

Syn wrote:
Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source code would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.

/**
* Construct and initialize structure from a 14-digit String timestamp. If
* the argument is too short, or specifies an invalid timestamp, cleanup
* will be attempted to create the earliest legal timestamp given the input.
* @param dateStr from which to set date
*/
public Timestamp(final String dateStr) {
setDate(dateStrToDate(dateStr));

full source here:
[You must be registered and logged in to see this link.]

Is there an emulator I can run this code in?

by jeanmonroe 20.06.15 15:54

Oh, come 'back', Nullah.

Mrs Boss (me) will 'sing you to me'

"Somewhere, over the rainbow, waayyy up high"

We ain't all the 'same', be pretty damn 'boring' if we were, wouldn't it?

friends

----------------------------------------------------------------------------
"If it's a WaybackMachine error, then their credibility is screwed. But, if that date is accurate, it's dynamite!" (Almostgothic)

by Guest 20.06.15 16:05

Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source code would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.

Exactly.

I don't like "new" members with a mission to educate us on a VERY hot topic telling us that they know it's impossible because well... they say so.

So excuse my Yorkshire bluntness, why mess about?

by sallypelt 20.06.15 16:07

BlueBag wrote:
Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source code would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.
Exactly.

I don't like "new" members with a mission to educate us on a VERY hot topic telling us that they know it's impossible because well... they say so.

So excuse my Yorkshire bluntness, why mess about?

Sleeping spiders, spider crawls, webscraping...............it's all getting too much for me................HELP!!!!

by Guest 20.06.15 16:08

SixMillionQuid wrote:Just chiming in briefly.

1. If this was a 'glitch' why would an automated system pick a precise date time of 30/04/2007 11:58:03? It makes no sense to make up a date before the the web page became accessible to a webcrawler.

2. If this was a 'glitch' then other pages archived around the same time would show similar dates and times. Wayback should be able to confirm this.

Yes - and provide examples.

A point I made hundreds of posts ago.

by sallypelt 20.06.15 16:10

BlueBag wrote:
SixMillionQuid wrote:Just chiming in briefly.

1. If this was a 'glitch' why would an automated system pick a precise date time of 30/04/2007 11:58:03? It makes no sense to make up a date before the the web page became accessible to a webcrawler.

2. If this was a 'glitch' then other pages archived around the same time would show similar dates and times. Wayback should be able to confirm this.
Yes - and provide examples.

A point I made hundreds of posts ago.

If what my limited IT knowledge is telling me, then what I am understanding is, if a person or company is designing a web page, there are programmes available which can capture that information WITHOUT it having hypertext before it goes live. Please techies, can you tell me that I am barking up the wrong tree?

by Guest 20.06.15 16:13

Syn wrote:
Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source code would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.

/**
* Construct and initialize structure from a 14-digit String timestamp. If
* the argument is too short, or specifies an invalid timestamp, cleanup
* will be attempted to create the earliest legal timestamp given the input.
* @param dateStr from which to set date
*/
public Timestamp(final String dateStr) {
setDate(dateStrToDate(dateStr));

full source here:
[You must be registered and logged in to see this link.]

I'm a professional java programmer with 15 years java experience.

You've told us nothing here except what the method comment says.

How does this relate to the issue at hand?

by Syn 20.06.15 16:13

Richard D. Hall wrote:
Syn wrote:
Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source code would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.

/**
* Construct and initialize structure from a 14-digit String timestamp. If
* the argument is too short, or specifies an invalid timestamp, cleanup
* will be attempted to create the earliest legal timestamp given the input.
* @param dateStr from which to set date
*/
public Timestamp(final String dateStr) {
setDate(dateStrToDate(dateStr));

full source here:
[You must be registered and logged in to see this link.]

Is there an emulator I can run this code in?

Java Virtual Machine https://java.com/en/download/

by Guest 20.06.15 16:16

sallypelt wrote:
BlueBag wrote:
SixMillionQuid wrote:Just chiming in briefly.

1. If this was a 'glitch' why would an automated system pick a precise date time of 30/04/2007 11:58:03? It makes no sense to make up a date before the the web page became accessible to a webcrawler.

2. If this was a 'glitch' then other pages archived around the same time would show similar dates and times. Wayback should be able to confirm this.
Yes - and provide examples.

A point I made hundreds of posts ago.
If what my limited IT knowledge is telling me, then what I am understanding is, if a person or company is designing a web page, there are programmes available which can capture that information WITHOUT it having hypertext before it goes live. Please techies, can you tell me that I am barking up the wrong tree?

It is possible to search and copy (scrape) a server by looking for new publicly accessible objects that are not linked.

You can interrogate a directory for what's in it - your browser does it all the time with your cookies.

by Syn 20.06.15 16:16

BlueBag wrote:
Syn wrote:
Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source code would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.

/**
* Construct and initialize structure from a 14-digit String timestamp. If
* the argument is too short, or specifies an invalid timestamp, cleanup
* will be attempted to create the earliest legal timestamp given the input.
* @param dateStr from which to set date
*/
public Timestamp(final String dateStr) {
setDate(dateStrToDate(dateStr));

full source here:
[You must be registered and logged in to see this link.]
I'm a professional java programmer with 15 years java experience.

You've told us nothing here except what the method comment says.

How does this relate to the issue at hand?

Wayback appends java into the html during archival. The above code calculates date and timestamp.

This should be useful to those who want to have look through it http://www.javased.com/index.php?source_dir=wayback/wayback-core/src/main/java/org/archive/wayback/util/Timestamp.java#

PS not sure why you are wary of me. I only joined here as I was able to answer a question about orphan pages. I am a staunch 'anti' Hate that word, prefer pro Madeleine but hey ho. If you want to see what my thinking is on McCann case I use twitter as @Syn0nymph but if you don't then here's my greptweet http://greptweet.com/u/syn0nymph/syn0nymph.txt

PS Nuala was right about orphan pages

'Orphan pages: If there are no links to your pages, the robot won't find it (the robots don't enter queries in search boxes.)'

This above is from the horses mouth http://web.archive.bibalex.org/web/policies/faq.html

by Guest 20.06.15 16:18

Syn wrote:
BlueBag wrote:
Syn wrote:
Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source code would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.

/**
* Construct and initialize structure from a 14-digit String timestamp. If
* the argument is too short, or specifies an invalid timestamp, cleanup
* will be attempted to create the earliest legal timestamp given the input.
* @param dateStr from which to set date
*/
public Timestamp(final String dateStr) {
setDate(dateStrToDate(dateStr));

full source here:
[You must be registered and logged in to see this link.]
I'm a professional java programmer with 15 years java experience.

You've told us nothing here except what the method comment says.

How does this relate to the issue at hand?
Wayback appends java into the html during archival. The above code calculates date and timestamp.

Yes, so what?

How does that relate to the issue at hand?

by **Richard D. Hall** 20.06.15 16:19

Syn wrote:
BlueBag wrote:
Syn wrote:
Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source code would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.

/**
* Construct and initialize structure from a 14-digit String timestamp. If
* the argument is too short, or specifies an invalid timestamp, cleanup
* will be attempted to create the earliest legal timestamp given the input.
* @param dateStr from which to set date
*/
public Timestamp(final String dateStr) {
setDate(dateStrToDate(dateStr));

full source here:
[You must be registered and logged in to see this link.]
I'm a professional java programmer with 15 years java experience.

You've told us nothing here except what the method comment says.

How does this relate to the issue at hand?
Wayback appends java into the html during archival. The above code calculates date and timestamp.

Is it bug free? - or does it occasionally get the date wrong?

by jeanmonroe 20.06.15 16:22

sallypelt wrote:

"Sleeping spiders, spider crawls, webscraping...............it's all getting too much for me................HELP!!!!"
---------------------------------------------

KA-CHINGGGGGGGG!

If we 'scrape' all those 'webs' away, there'll be no place for them sleeping sipders, crawly spiders to 'hide', geddit?

38 'pages' to 'reach' a 'solution' that was staring us in the face, all along!

PAH!

ps: Sally, you'll have to do the 'spider webs scraping'............i is 'well scared' of 'spiders', innit?

by Guest 20.06.15 16:25

Richard D. Hall wrote:
Syn wrote:
BlueBag wrote:
Syn wrote:
Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source code would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.

/**
* Construct and initialize structure from a 14-digit String timestamp. If
* the argument is too short, or specifies an invalid timestamp, cleanup
* will be attempted to create the earliest legal timestamp given the input.
* @param dateStr from which to set date
*/
public Timestamp(final String dateStr) {
setDate(dateStrToDate(dateStr));

full source here:
[You must be registered and logged in to see this link.]
I'm a professional java programmer with 15 years java experience.

You've told us nothing here except what the method comment says.

How does this relate to the issue at hand?
Wayback appends java into the html during archival. The above code calculates date and timestamp.

Is it bug free? - or does it occasionally get the date wrong?

Your original point is valid.

You would need to see all the code not just this.

This has a "make do" function if garbage is put in when creating a timestamp.

Did it ever have to "make do"?

I often code for "just in case".

by rustyjames 20.06.15 16:26

Wow - really struggling to keep up with this thread so this is a slightly random collection of things from the previous dozen or so pages and will no doubt be out of date by the time I post it!!

- First we need to remember the Internet Archive is a not for profit organisation whose primary objective is creating a "library" of digital information, since in this day and age where everything isn't documented on paper much of our history could be lost. Legal use is a by product.

- As per an earlier post I agree with Richard that it's impossible for anyone other than the WBM to know how their software works

- I also agree that the timestamp from 30th April is the most important thing to understand. As I said many pages back, database updates can go wrong with multi-threading, lack of locking etc, but effectively timestamping sequential logged events it's less likely. And I'm sure the WBM servers all sync their times with NTP.

- Scraping vs Crawling - not much of a difference really. Scraping generally refers to programatically grabbing info from a specific website for use elsewhere. For example grabbing prices from a website where there wasn't an official way for programmers to get that info. Crawling refers to following links to many websites as you would if you clicked every link on a page you knew and then followed every link on the pages you end up on and so on. A search engine does this and indexes everything it finds so we can then use it to search for things.

- I'm sure it's not the case here, but I can think of a couple of scenarios where an item not explicitly linked might be found. The first is if there is a link to a folder not containing an index file - then the contents of the folder (dependant on security such as .htaccess) would be served up and anything added into it would be visible. I don't know if they do it, but the other would be if a crawler looked for standard files such as index.html, index.asp etc.

- I disagree with the simplification of folders being created containing snapshots of the site. The vast amount of what the WBM archives will never be retrieved and I'm certain they won't be storing everything each time or pre-creating the pages, (they may well be caching it though). The URLs just drive a dynamically created page using the info beyond the /web/ part , (we know that anyway from the footer in the pages), with URLs rewriting of internal links etc. (see earlier post on the heavy use of HTTP 302 "Temporarily Moved" as well). Those used to debugging stuff can watch it in the network section of the F12 debugging in a browser.

- Don't know about the Norwich kit thing but the page may have had links to the kit that were not set up to actually render on the page, or more likely from the page it was possible to see where links for images were served up from and he grab them from there.

by Syn 20.06.15 16:28

BlueBag wrote:
Syn wrote:
BlueBag wrote:
Syn wrote:
Richard D. Hall wrote:In order to understand what the WBM does, it is necessary to have a print out of their servers source code, which was running on 30 April 2007. This is something that WBM probably have archived somewhere and could look at. As far as I am aware nobody other than them have visibility of this code, therefore only they know exactly (and I mean EXACTLY) how their crawler functions. If you were to get the code and do a code walkthrough I suspect the code will show that their systems seeks out new webpages, and newly updated webpages. When it finds a new webpage (such as mccann.html) on the ceop website, it will invoke code to copy that page onto a folder on its server. In order to copy the new file it will have a function to create the new folder. This function I am sure will use the systems real time clock to construct the folder name based on the current time and day. Once the folder is made on their server, the file is then copied. Their system also ammends the file by adding a date and time stamp within the file and also some other information it might need at a later time. My description here is a general one because I do not have a copy of their source code. I think it is unlikely that a walkthrough of their source code would reveal that it is possible to create a folder with a date which pre-dates the date when a new file is discovered.

/**
* Construct and initialize structure from a 14-digit String timestamp. If
* the argument is too short, or specifies an invalid timestamp, cleanup
* will be attempted to create the earliest legal timestamp given the input.
* @param dateStr from which to set date
*/
public Timestamp(final String dateStr) {
setDate(dateStrToDate(dateStr));

full source here:
[You must be registered and logged in to see this link.]
I'm a professional java programmer with 15 years java experience.

You've told us nothing here except what the method comment says.

How does this relate to the issue at hand?
Wayback appends java into the html during archival. The above code calculates date and timestamp.
Yes, so what?

How does that relate to the issue at hand?

Wayback has in the past had issues using java method as described below.

[You must be registered and logged in to see this link.]

Back laters :)

by rustyjames 20.06.15 16:31

Going to struggle for time to read it, but the wiki could be very interesting - [You must be registered and logged in to see this link.]

by Guest 20.06.15 16:32

I disagree with the simplification of folders being created containing snapshots of the site.

I don't think anyone said that.

It is snap shots of bits of the site though.

The BBC website that I showed earlier is a good one to pick it apart - and the time stamp was 100% accurate because the BBC page had the same time embedded in it's top news item.

by Guest 20.06.15 16:34

Nuala wrote:But Nuala has already said it's impossible so it must be impossible because Nuala knows exactly how the Way BAck machine program worked in 2007 and that's why he/she signed on urgently to tell us.

Still no real explantation from the real people.

Right, that's it, I'm gone.

I'm not putting up with such rudeness when all I'm doing is trying to help.

========================================

Very sorry to see you go; please don't

Tempers here do get frayed from time to time

friends

by rustyjames 20.06.15 16:37

This 2007 document from Brad Tofel who was a software engineer at the Internet Archive at the time could be even more interesting to understand how it works...

[You must be registered and logged in to see this link.]

by MRNOODLES 20.06.15 16:37

Is it me or am I paranoid. But have we just be corralled down the 'Java' road. When that was what CEOP claimed in the first instance.

You are the CEOP web developer and I claim my £5

by jeanmonroe 20.06.15 16:39

Portia wrote:
Nuala wrote:But Nuala has already said it's impossible so it must be impossible because Nuala knows exactly how the Way BAck machine program worked in 2007 and that's why he/she signed on urgently to tell us.

Still no real explantation from the real people.

Right, that's it, I'm gone.

I'm not putting up with such rudeness when all I'm doing is trying to help.
========================================

Very sorry to see you go; please don't

Tempers here do get frayed from time to time

-----------------------------------

NO, THEY F**KING DON'T!.................Oooppppss!

laughat

Latest topics