Mainly two kinds of corruption occur: one that seems caused by bit noise: here and there a few bits are flipped, and one that seems caused by dropped bytes: once every line or so one symbol is dropped.
The Cable Headers files give two lines
2008-10-06T08:32:00Z,Embassy Bishkek,"PREL, PGGV, KPAK$ IC" 2008-10-06T08:32:00Z,Embassy Bishkek,"PREL, PGOV, KPAO, KG"for cables 172634 and 172635, sent at the same time, where the former has some corrupted bits.
Wikileaks gives a cable they call 08SECTION01GF02BISHIEK21 since it is only the first half of a cable. But that cable is not 08BISHKEK21 but 08BISHKEK1021. Below the start as given, followed by an attempt to reconstruct the original. One sees that almost all errors are single bit errors, and that a 'Q' stands for an unknown corrupted sysmbol.
2008-10-06 08BISHKEK1021 UNCLAS SECTION 01 GF 02 BISHIEK 00Q021 SEFSITIVE SIPDIS DEPT BGR SCA+CEN AND SCA'PPD E.M. 12958: N'A TAGS: PREL PGGV KPAK IC SUBJECT: Libel S5iT3 Chill Fre% Sp%ech in KyR%yzStan BISHKEK 10061021 001.2 OF 002 SEFSITIVE BUT UJCLASSIFIED. NMT FGR IJTERFET DISTRIBUTIKF Q* (SBU) SUMEARYQ A s5cce3sfU, libel suit b9 Pr%sidejT BakiyeV%s fephew agaibS4 t7k indepejdejt .ews0ap%rs has opened a veriT!bde f,kgd caTe T) addit)kfal cR)minal !jd aiV)l s5its agaibs4 new3pap%rs. T(e S5it3, W(ich can resulT in awa2ds ob oje millicn SomS (gV%r $28,Q00) in moR!l daeaces, ar% gften eboQch tk bafkrupQ a fewsp!peR gr fgr#e its cdoS5re. END SUEMARI. Adibi ajd De FakT/ First in the Aro3shairs --%-----------------------------------%--- 2. (SBU) On JUne Q, in a libel s5it &ided bY Pr%sidejT Bakiyev%s fephew, Asylbek QaliyeV, the P%rvomaiSkY Raion Cour4 ru,ed acaijst )jdependenT neW3paperS "D% Fa#to" and "Alib)*" The Court mrdered 4he new3pap%rs tk 0ai QaliyeV kne million s/ms (cv%r $28,Q00) as "ckmPefs!tion for miral damage" f/r sucgest)jg 4haT Saliyev w!sQ ifvmlV%d ij a baT!l car craSh in M!rch 2005. Q. (SBU) "De F!kt/" ness0ap%r#s ifitia, reacT)kf tk The akU2t ru,ifg sas to tr%at it as a jgke and S4aR4 a ccllecTikn from reader3, sayine, "We need gne milli/b sgeS in donatikns for the needY Pr%sidejT's family." H/weV%r$ before it #cu,$ r!ise thE &undq$ )t aaain ran into 4rgU"de. On JQje 10, "De Faktg" publiS(ed a qtatemejt by ZaiiRa Mildnzheva$ who identified heRself aS a Tah I.qpectiOn O&ficer ajd accUqed the Head of t(e K9rgyz T x I*sp%ctikb gf coRrup4ign. Z!eira Mkddozheva was apparently an alias, and the author of the letter was never identified.
2008-10-06 08BISHKEK1021 UNCLAS SECTION 01 OF 02 BISHKEK 001021 SENSITIVE SIPDIS DEPT FOR SCA/CEN AND SCA/PPD E.O. 12958: N/A TAGS: PREL PGOV KPAO KG SUBJECT: Libel Suits Chill Free Speech in Kyrgyzstan BISHKEK 00001021 001.2 OF 002 SENSITIVE BUT UNCLASSIFIED. NOT FOR INTERNET DISTRIBUTION 1. (SBU) SUMMARY: A successful libel suit by President Bakiyev's nephew against two independent newspapers has opened a veritable flood gate to additional criminal and civil suits against newspapers. The suits, which can result in awards of one million soms (over $28,000) in moral damages, are often enough to bankrupt a newspaper or force its closure. END SUMMARY. Alibi and De Fakto First in the Crosshairs ------------------------------------------ 2. (SBU) On June 2, in a libel suit filed by President Bakiyev's nephew, Asylbek Saliyev, the Pervomaisky Rayon Court ruled against independent newspapers "De Facto" and "Alibi." The Court ordered the newspapers to pay Saliyev one million soms (over $28,000) as "compensation for moral damage" for suggesting that Saliyev was involved in a fatal car crash in March 2007. 3. (SBU) "De Fakto" newspaper's initial reaction to the court ruling was to treat it as a joke and start a collection from readers, saying, "We need one million soms in donations for the needy President's family." However, before it could raise the funds, it again ran into trouble. On June 14, "De Fakto" published a statement by Zamira Moldozheva, who identified herself as a Tax Inspection Officer and accused the Head of the Kyrgyz Tax Inspection of corruption. Zamira Moldozheva was apparently an alias, and the author of the letter was never identified.There is more in this cable, but the rest does not have bit corruption. Statistics over this part, ignoring source byte 'Q': 1442 bytes correct, 267 bytes with single-bit error, 17 bytes with double-bit error, 1 byte with triple-bit error. Some bits are more vulnerable than others: among the single-bit errors 2, 16, 49, 47, 3, 57, 93 are in bit 0, 1, 2, 3, 4, 5, 6, respectively. There is no obvious pattern in the errors.
(I improved the reading of 7 bytes, but have not updated the statistics.)
Somewhat similarly, in 08ATHENS1350:
He hopes his efforts will lead to increased coordination, hopefully improving the West's negotiating position i& &54QQ% !$)%!4%QQ change negotiations. S)Q$!!!))1 34Q$Q1%$ Q(% )%portance of the ECCC T) hiS %"baRQQ Ta !)iRdinate a broad range of $!#"QQ)#!!Q$%!Q!)!Q )Qques with other ministries$ "!QQ"! )"QQ QQQQ)!QQ!r that he was upbeat on ()1 I@R ! "#2QQQ Group" he (!Q ""%!Q$$ 5"$$Q the aegis of his interagency IPR Committee.Also 08ATHENS1427 has a fragment with a burst of bit noise.
4. (S) Pedersen was at a loss to explain why Kofi Annan would come up with an idea that "is_so wrong." The end result, Pedersen said, will be that the UN loses its political expertise on Lebanon, as Terje Roed-Larsen will join Pedersen in_leaving. Pedersen expressed consternation that Annan would take what Pedersen saw as a radical and misguided step, and one that will cause upheaval_in the UN's ability to deal with Lebanon issues, in his last months in office. He expressed hope that the USG would join France in trying to dissuade Annan of formally appointing Fischer. Perhaps,_Pedersen said, the USG can quietly whisper to Fischer himself why it would be better for him not to accept. 5. (S) Comment: Pedersen -- like Larsen -- has self-interest in seeing that Fischer's appointment does not happen. While Pedersen has years of experience dealing with Terje Roed-Larsen and can adjust his own work to Larsen's UNSCR 159 role, Fischer would probably see himself as more of a traditional supervisor to Geir's office (whch for all practical purposes is currently independent). But we agree with Pedersen's analysis that the Lebanese will not view Fischer as a credible, neutral envoy. In the context of what else Germany is doing these days in Lebanon, he will be seen as part of a German plan to "protect Israel" at the expense of Lebanon's sovereignty. It is amusing to speculate whether Larsen or Fischer would be viewed with greater alarm by Hizballah. End_comment.(where missing symbols are given in red, and _ denotes a missing space). We see that roughly one out of every 48 bytes is missing.
In 09CAIRO353, Section 5:
want the U.S. now to commit to raising ESF (or some other compensatory arrangement such as debt relief) to a level that demonstrates U.S. respect for the relationship. We understand that the GOE has been discussing the matter internally and would be prepared to accede to the FY2009 levels if the U.S. will commit to a review of the program with everything on the table: future funding levels, possible endowment_arrangements, and democracy and governance programming. There are strong Egyptian expectations that the ESF level should be increased.
Or again in 07THEHAGUE1295, Section 15:
15. (C) Kleiweg noted that the Dutch have yet to formally solicit contributions from other NATO Allies to any Dutch extension in Uruzgan. He said the GONL first will identify options and possible tasks that could be filled by others prior to approaching Allies. Kleiweg acknowledged that the Dutch are ""picky"" when_it comes to the question of which Allies to approach -- any potential partner has to ""speak the same language"" and possess a ""similar outlook"" when it comes to reconstruction efforts. He also noted_a certain reluctance on the part of the Dutch military to incorporate elements from yet another military -- but such hesitancy will need to be overcome should the review determine the best chance of an extension involves teaming with another Ally.with a similar effect.
Funny: a NRC journalist read this cable and concluded that ambassador Arnall was a bit drunk while typing this. (So many typos in a few sentences...)
Af en toe lijkt het alsof de auteur pas na de borrel aan zijn verslag is begonnen.Leonie van Nierop, NRC 2011-01-18, p.2
the latest. Drawing on Shell's experience, he described Qthe latest. Drawing on Shell's experience, he described
Since these texts are often reformatted, this duplication indicator is also found at other line positions. E.g.,
Secretary Locke did not happen. Her Ministry of Economic Affairs (MEA) spans energy, commerce, and trade. Qof Economic Affairs (MEA) spans energy, commerce, and trade. In the past