Tech | Steve Woods

Firefox Focus – first impressions

02/05/2023 Steve WoodsBristol, Open Source, Privacy, Security, Tech

Your ‘umble scribe is a great fan of the free and open source Firefox web browser and has been using the desktop version since version 0.x many years ago. One of its major attractions has been its emphasis on security and privacy.

Until recently it was also the default browser on my smartphone, until I discovered Firefox Focus. Firefox Focus is a free and open-source privacy-focused mobile browser based on Firefox which is available for Android and iOS devices. First released in December 2015, it was initially a tracker-blocking application for mobile iOS devices, but was developed into a minimalistic web browser shortly afterwards.

According to Mozilla, Firefox Focus is a dedicated privacy browser with automatic tracking protection. meaning web pages load faster and your data stays private. It’s also easy to delete history, passwords and cookies, so advertisers and other ne’er-do-wells don’t follow you around online. Just tap the erase button on the search field and all that data is gone. Tracking protection is also very strong. The browser blocks a wide range of common trackers by default, including social trackers and those sticky ones that come from things like Facebook ads.

After using Firefox Focus for one week, I can say I’m impressed with the way it works. Although it required me to learn how to use tabbed browsing (hint: hold down a link in your search results and a menu appears, offering the option to open the link in a new tab. Ed.), once that was cracked, I was away. As for fast page loading, that’s not disappointing either, even on notoriously slow-loading sites, like that of Bristol City Council, which still seems to be powered by a horse turning a shaft in the basement of the Counts Louse (which some call City Hall. Ed.). 😉

If you value your privacy and security, I’d recommend Firefox Focus on your mobile device.
French Customs censured for illegal retention of personal data

21/04/2023 Steve WoodsPrivacy, Security, Tech

French IT news site Le Monde Informatique reports that the French Customs authorities have been sent a formal notice by the CNIL, France’s data privacy regulator, in respect of an illegal data file containing the details of more than 45,000 people, including copies of identity documents and records of criminal offences.

Businesses are not the only organisations with which the CNIL has found fault for holding illegal files containing personal data. Public sector organisations can also fall foul of the law.

The French Customs authorities, which come under the control of the Ministry for the Economy have been caught red-handed following a report in respect of Customs’ file used for recording information about vessels and their crews which is known as SIRENE. Intended to identify all the people checked at sea or in port in order to combat fraud, this system was in fact developed and implemented with no legal basis and not in accordance with the law, according to the CNIL

Checks were carried out by Customs’ Channel-North Sea-Atlantic coastguard service and inspections revealed that recourse to this system did not comply with France’s Data Protection Act. This data system actually lists information about the vessels checked and their passengers, including personal information such as marital status, address, occupation and copies of identity documents, as well as criminal convictions (drug trafficking, counterfeiting, off-the-books employment, failure to co-operate, sexual assault, possession of illegal weapons, intentional homicide and murder).

6 months to comply or be fined

All told, the details of 45,793 persons – including 392 minors – are included in the SIRENE file. “The creation and use of the SIRENE file are not provided for by any legislation (for example a law or a decree). In addition, the CNIL has not received a request for an opinion concerning its implementation, in violation of the Data Protection Act (articles 87 and 89, the CNIL explained. Other grievances have also been lodged against the Ministry for the Economy, such as the failure to send an impact assessment in respect of the protection of personal data and the lack of a clear distinction between the data of the different categories of persons concerned. or the fact that the latter were not made aware that their data had been included.

Following the CNIL’s formal notice, the Ministry for the Economy and Customs have 6 months to comply otherwise a penalty could be issued.
Seriously

06/04/2023 Steve WoodsLanguage, Security, Tech
The language used in official responses to news stories seems to have been rigid and formulaic in recent times, particularly amongst those organisations within or linked to the public sector.

Today’s edition of The Register reports that ACRO, the UK’s Criminal Records Office was taken offline due to a security breach. The site currently displays a holding page blaming ‘technical issues‘, a fine example of misleading bureaucratic language.

This is the site’s holding page as this post is published.

El Reg notes that manages ACRO people’s criminal record information, running checks as needed on individuals for any convictions, cautions, or current prosecutions. It with British police and businesses, as well as exchanging this data with other countries, particularly where people wish to move or emigrate to another country and a certificate of good behaviour is required from the British police. ACRO has access to data from the Police National Computer via an information sharing agreement with the Cabinet Office.

The data typically handled by ARCO includes name and address history, extended family information, a new foreign address, legal representation, passport information, photo and data PIN cautions, reprimands, arrests, charges or convictions.

Earlier this week, ACRO emailed users to inform them that it had “recently been made aware of a cyber security incident affecting the website between 17th January 2023 and 21 March 2023“, adding that “we have no conclusive evidence that personal data has been affected by the cyber security incident; however it is only right that we inform you of the situation. We are very sorry that because of your interaction with ACRO your data could have been affected, and we are working tirelessly to resolve this matter.”

The message went on to say that “robust measures” had been taken as soon as the breach was discovered. It won’t be the first time that pulling the plug on a website has been described by a public sector organisation spokesperson as “robust”, If your systems were truly “robust”, taking the site offline would not have been necessary.

After intoning the “robust” mantra, ARCO then goes on to say: “We take data security very seriously and will ensure that the matter is fully investigated…. Translating this into plain English, this means “Oh dear! We’ve been caught out!”

The fact that ARCO had not taken data security “very seriously” is clearly highlighted by two facts:
- Firstly, ARCO did not notice crooks were gaining access to its computer systems for more than two months; and
- Secondly, it has now freely admitted that it is going to take steps to find out how the breach happened and prevent its reoccurrence. A clear case of that old adage of shutting the stable door after the horse has bolted.
The public sector relies heavily on public trust to do its work. If it really does want to be taken seriously, tough measures need to be taken and implemented, not just for IT security, but in connection a very ancient and fundamental idea: that of honesty.
Happy 25th, curl!

22/03/2023 Steve WoodsLinux, Open Source, Tech

Version 8 of the curl command line too has been released, German It news website reports. This coincides with the software’s 25th birthday.

The release of a new major version (8.0O of curl (Client for URLs) has been released just in time for the software’s 25th birthday. The data transfer command line tool has barely changed. The new release has far more to do with publicising the birthday of the software and its libcurl program library. This was explained by curl’s initiator and maintainer Daniel Stenberg when announcing the release. In moving to a version 8Stenberg also wanted to avoid ending up with a curl version seven with point releases running to three figures (7.xxx).

Little has changed within curl itself with this release: 8.0 is just the first release of curl that no longer runs on systems without working 64-bit data types, as can be gathered from the release notes. Otherwise, the new version contains 130 bug fixes, including six vulnerabilities of which Stenberg classifies five as “low” and one as “medium“. Furthermore, there are rewards ranging from $480 to $2,400 for those who successfully squash curl’s bugs.

To celebrate the release, some of the project’s figures have also been released. There have been 215 releases, whilst 41 contributors (of whom 23 were new) collaborated on version 8.0. A total of 2,841 persons have contributed to curl’s code; mostly only once, as Stenberg comments in his Youtube video.

Curl itself is a very popular command line tool for sending and receiving data with URL syntax, whilst libcurl is a transfer program library which handles most internet protocols and is used in many third party applications.
Czech government using open source web analytics

17/03/2023 Steve WoodsGDPR, Open Source, Privacy, Security, Tech

Joinup, the EU’s open source news site, reports that the Czech Republic is to begin using the Matomo open source web analytics tool on the Czech citizen portal and gov.cz websites, where it will replace Google Analytics.

This change will ensure that the data by the sites collected will stay within the EU and, as the Czech administration will be using its own instance of Matomo, it will retain full control of the records.

The change was triggered by an open letter sent by the Czech the digital freedom watchdog luridicum Remedium after it noticed the Czech state vaccination system website was using Google Analytics during the COVID-19 crisis. The Czech Data Protection Authority and public sector strategic partner NAKIT then pursued the matter and replaced Google Analytics with Matomo on Czechia’s Ministry of Health website. This move later led to further action and the country will continue following this trend on public sector websites.

Previously named Piwik, Matomo has been in development since 2007 and is presently deployed on 1.4 million websites, including those of NASA, the European Commission, the United Nations and Amnesty International.

The Czech decision to choose Matomo follows those of other European countries seeking to keep control of their citizens’ data. Last year the French and Austrian data protection authorities determined that Google Analytics was not compliant with EU data privacy standards, in particular because Google’s data transfers to the United States are contrary to the EU’s General Data Protection Regulation (GDPR).
LibreOffice Base Guide – now in Czech

16/03/2023 Steve WoodsLanguage, Open Source, Open Standards, Tech

The blog of The Document Foundation, the organisation behind the free and open source LibreOffice productivity suite, reports that the user guide for Base, the suite’s database development and administration tool for relational database management systems has now been translated into Czech.

Czech LibreOffice community member Zdeněk Crhonek (aka “raal”) writes as follows:

The Czech team translated the LibreOffice Base Guide 7.3 – and it’s now available on the documentation page. Our team consists of three translators: Petr Kuběj, Radomír Strnad and Zdeněk Crhonek, along with localized screenshot maker Roman Toman, and Miloš Šrámek, who prepared machine translations.

Learn more about or join the LibreOffice Documentation project.
LibreOffice & Nextcloud for EU Institutions

27/02/2023 Steve WoodsOpen Source, Politics, Security, Tech

EU data protection authorities have negotiated a contract for the use of Nextcloud and LibreOffice Online in EU institutions. They are now testing the solutions, German IT news heise reports.

Data protection-friendly alternatives

It was announced last Wednesday that the European Data Protection Supervisor Wojciech Wiewiórowski and his team have begun testing both solutions this month. In coming months they want to examine “how these can tools support EU day-to-day work“. This pilot phase is part of a larger IT reflection process that the EDPS already started last year aimed at encouraging EUIs to consider alternatives to large-scale service providers to ensure better compliance with Regulation (EU) 2018/1725.

By procuring the Open Source Software from one single entity in the EU, the use of sub-processors is avoided. In doing so, the EDPS avoids data transfers to non-EU countries such as the USA and allows for more effective control over the processing of personal data.

According to Mr Wiewiórowski, “Open Source Software offers data protection-friendly alternatives to commonly used large-scale cloud service providers that often imply the transfer of individuals’ personal data to non-EU countries. Solutions like this may therefore minimise reliance on monopoly providers and detrimental vendor lock-in. By negotiating a contract with an EU-based provider of cloud services, the EDPS is delivering on its commitments, as set out in its 2020-2024 Strategy, to support EUIs in leading by example to safeguard digital rights and process data responsibly“.

Microsoft Office in the sights

Mr Wiewiórowski has already examined the contracts which EU institutions have with Microsoft and reached the conclusion in 2020 that the data processing purposes when using Windows or Microsoft Office had been defined far too openly. Processing contractors were not adequately audited and data could be transferred too easily by EU institutions to countries outside the Union. At the time, he demanded that Microsoft should only retain user information within the EU. The roles of all those involved with all their rights and obligations must be clearly regulated. Furthermore, Users should look around for alternatives that “enable higher data protection standards“.

The EDPS started further investigations into the use of Microsoft and Amazon cloud services by EU institutions. These entailed the use Microsoft Office 365 by the EU Commission. According to Wiewiórowski many contracts were concluded prior to the “Schrems II Judgment” and had to be examined in the light of the European Court of Justice case law.
Content liability: Big Tech squares up to Uncle Sam

27/01/2023 Steve WoodsPolitics, Privacy, Social Media, Tech

Following the announcement anti-trust action by the United States Department of Justice along with the Attorneys General of California, Colorado, Connecticut, New Jersey, New York, Rhode Island, Tennessee, and Virginia against Google, Meta (owners of Facebook and Instagram), Microsoft and Twitter have all made statements seeking to defend their actions.

In their legal opinions, the big US tech giants, including Microsoft, Meta and Twitter, are warning the Supreme Court against amending Section 230 of the Communications Decency Act (CDA). This would enable actions against content recommendation algorithms, French IT news site Le Monde Informatique reports.

One week after Google’s filing of a defence statement with the US Supreme Court warning that amending Section 230 of the Communications Decency Act (CDA) “would upend the internet“, several companies including Twitter, Meta and Microsoft, have filed their own legal opinions. They support Google’s argument that a restriction of the law could have disastrous consequences for the content editors. By virtue of the 1996 CDA, the companies are shielded from liability for content posted by their users, including comments, criticism and advertising.

US Supreme Court.
Image courtesy of Wikimedia Commons and UpstateNYer

However, the Supreme Court has been asked to examine whether Section 230 was still pertinent and appropriate, given that it was promulgated before the internet became part of everyday life. The law was subject to a minute before the suit filed by the family of Nohemi Gonzalez, a 23 year-pld US citizen killed in Paris during the 13th November 2015 terrorist attacks claimed by ISIS. The Gonzalez family asserts that the algorithms should be regarded as editorial content not covered by the immunity from liability granted by Section 230 and thus Google’s YouTube subsidiary has violated the US Anti-Terrorism Act (ATA) when its algorithms have recommended ISIS-linked content to users. The Supreme Court is set to hear oral arguments in the case on 21st February next.

Criticisms of the protections of Section 230 for websites

Both Democratic and Republican members of Congress have criticised the protections provided for by the law. The Republicans believe that those in respect of liability make websites take partial decisions regarding content removal, whilst the Democrats would like the same sites to take more responsibility as regards moderation. In a statement President Biden has stated that his administration would support the position that Section 230 protections should not apply to recommendation algorithms. In its petition of 19th January, Microsoft asserts that if the Supreme Court makes amendments to Section 230, it would “strip these digital publishing decisions suit—and it would do so in illogical ways that are inconsistent with how algorithms actually work.“.

The company added that any decision aimed at restricting the law “thereby expose interactive computer services to liability for publishing content to users whenever a plaintiff could craft a theory that sharing the content is somehow harmful“. In its own petition Meta stated that the plaintiffs’ argument is “deeply flawed from a legal point of view”; by interpreting Section 230 as a means of protecting sites from liability for content posted by its users whilst removing protection from content “ignores the way in which the internet works“. The company continued by describing the plaintiffs’ position as “regrettable from a practical point of view” and by stating that a ruling in their favour would ultimately prompt “online services to remove important, provocative and controversial content on matters of general interest“.

Protection from liability essential for website operation according to Twitter

Twitter has said that the current interpretation of Section 230 “ensures that sites such as Twitter and YouTube can work in spite of the unfathomable amount of information they make available and the potential liability that might result from this“. Since Twitter’s acquisition by Elon Musk, the site has been criticised for having reinstated the accounts of people it previously banned, such as disgraced former president Donald Trump or alpha male par excellence and all-round amateur human being Andrew Tate who is currently under investigation in Romania for alleged human trafficking.

However, the review of several other high-profile cases will have to take place before the law is changed. Last week the Supreme Court was set to discuss its jurisdiction in two cases that challenge Texas and Florida laws prohibiting online platforms from removing certain political content. In addition, a Twitter vs. Taamneh case, which has many similarities with the Gonzalez vs. Google case, is due to oral pleadings on 2nd February. In this case Twitter, Facebook and YouTube are accused of having aided and abetted another attack claimed by Islamic State.
Another data protection fine for Meta

20/01/2023 Steve WoodsGDPR, Privacy, Social Media, Tech

After a record fine of €390 mn. at the start of January, the Irish Data Protection Commission is imposing a further fine of €5.5 mn. on Meta, this time for WhatsApp’s policy with regard to personal data under the GDPR, Le Monde Informatique reports.

Has been welcoming (in tax terms) to American IT companies, but is proving to be as very sensitive area for implementation of the GDPR. Meta has just experienced this once again with a fine of €5,5 mn. imposed by Ireland’s Data Protection Commissioner. This is the social network’s second fine in less than a month; on 4 January the same commission announced a record fine of €390 mn. on the personal data processing policy of Facebook and Instagram (posts passim).

In this instance it’s WhatsApp’s policy that is being censured following a complaint filed on 25 May 2018 – the date the GDPR entered into effect – by a German user. After this date the messaging service updated its general conditions of use and informed its users they had to click on “accept and continue” to indicate their consent. If they did not reply, they no longer had access to the service.As in the decision of 4th January, WhatsApp regards its data processing policy must be considered like a “contract” according to the GDPR (Article 6.1) concluded between the company and the user.

EDPB lays it on thick

The Irish Data Protection Commission investigated and drew up a draft decision which was submitted to the European regulators parties involved in this case. It proposed not imposing additional financial penalties. WhatsApp had already been fined €225 mn. in September 2021 for similar actions. However, the DPC pleaded for recognition of the contractual and thus legal nature of WhatsApp’s personal data policy – a position which caused an outcry from other data protection regulators.

The DPC approached the EDPB for a decision. It dismissed the legal basis of the contract and added an additional infringement of the transparency obligation. As a consequence, the Irish DPC is adding €5.5 mn. to the fine imposed on Meta, WhatsApp’s parent company.
GDPR"> Meta falls foul of GDPR

06/01/2023 Steve WoodsGDPR, Privacy, Social Media, Tech

Le Monde Informatique reports that Meta, the conglomerate that owns both Facebook and Instagram, has been fined a total of €390 for breaches of the EU’s General Data Protection Regulation (GDPR) in respect of both platforms’ personal data processing policy.

It has been a bad start to the year for Meta which has just been notified of a fine of €390 mn. by the Irish Data Protection Commission (DPC). The regulator is penalising the actions of Meta’s 2 subsidiaries, Facebook to the tune of €210 mn. and Instagram €180 mn. This decision concludes a case which started on 25 May 2018 (the date the GDPR entered into effect after 2 complaints had been filed – one by well-known Austrian privacy campaigner Max Schrems and the other by a Belgian citizen.

In this case Meta Ireland changed its general terms and conditions before the date of entry into effect of the GDPR, in particular “the legal basis on which it relied to legitimise its processing of users’ personal data (including behavioural advertising)”. To adopt this new policy, existing and recent Facebook and Instagram users were asked to click on the “I Accept” button on pain of no longer being able to access the platforms’ services. The questions then arose as to whether users had been forced to give their consent and if the “contract” concluded between Meta and its users conformed to Article 6 of the GDPR.

A fine increased by the EDPB

The debate was long and heated, including at European regulator level. As a matter of fact, the Irish DPC’s analysis did not meet with agreement from other European data protection authorities. For example, it considered the aspect of “forced consent” could not be upheld. Many authorities likewise thought the original Irish financial penalties too lenient. The European Data Protection Board (EDPB) was contacted to settle the matter and gave its decision on 5th December. It judged that “Meta Ireland was not entitled to invoke the legal basis of the “contract” as a legal based for its personal data processing for behavioural advertising purposes”.

It also demanded the fines proposed by the Irish regulator be raised. This is the second fine imposed on Meta in recent months by the CPD. Last November the American company was fined €275 mn. for so-called data scraping. In both cases, Meta still has the possibility of challenging the regulator’s decisions before the European judicial authorities.

Facebook and Instagram have now been given three months to bring their terms and conditions into line with the GDPR.