The definition covers code, model, and data information, with the latter being a contentious point due to legal and practical concerns. Mozilla, a long-time open source advocate, is partnering with OSI to promote openness in AI, advocating for transparency in AI systems.

The need to understand how AI systems work, so they can be researched, scrutinized and potentially regulated, is important to ensure the system is truly open source. Ayah Bdeir, senior strategic advisor on AI strategy at Mozilla, told SD Times on the “What the Dev?” podcast that AI systems are influenced by a number of different components – algorithms, code, hardware, data sets and more. 

As an example, she cited that there are data sets to train models, data sets to test, and data sets to fine tune, and this false sense of transparency leads organizations to claim their systems are open source. “When it comes to AI in traditional open source software, there’s a very clear separation between code that is written, a compiler that is used, and a license that is possessed. Each one of them can have an open license or a closed license and it’s very clear how each one of them applies to this concept of openness.” 

However, in AI systems, many components influence the system, Bdeir said. “This idea that if the code is open, that means their AI systems are open, which is not accurate.” This does not allow the fundamental reuse or study of the system that is required under an open source mentality, which is the actual four freedoms – use, study, modify and share, she explained.

“The open source AI definition by OSI is an attempt to put a real fine point on what open source AI is and isn’t, and how to have a checklist that checks for whether something is or isn’t, so that this ambiguity between claiming that something is open source or actually doing it is not is not there anymore,” she said. 

The debate over data information was among the most controversial in coming up with the definition, Bdeir said.  How do organizations that are training their models with proprietary data protect it from being used in open source AI? Bdeir explained there are schools of thought around data in particular. In one school of thought, the data set must be made completely open and available in its exact form for this AI system to be considered open source. “Otherwise,” she said, “you cannot replicate this AI system. You cannot look at the data itself to see what it was trained on, or what it was fine tuned on, etc. And therefore it’s not really open source.”

In another school of thought, where she said some of the more hands-on builders reside, making the data available is not realistic. “Data is governed by laws that are different in different countries. Copyright laws are different in different countries, and licenses on data are not always super clear and easy to find, and if you inadvertently or mistakenly distribute data sets that you have no rights to, you are liable legally.”

The OSI solution to this problem is to talk about data information. What OSI is requiring is data information, not the data in a data set. The wording, Bdeir said, says the organization must provide “sufficiently detailed information about the data used to train the system so that a skilled person can recreate a substantially equivalent system using the same or similar data.”

The post OSI officially releases its definition for Open Source AI appeared first on SD Times.

To learn about the findings of the report and what they mean, we spoke with Thabang Mashologu, VP of community at the Eclipse Foundation, on the most recent episode of our podcast, What the Dev?

Here is an edited and abridged version of that conversation: 

How did this survey come about? What made you want to study the impact of open source in these areas?

First of all, since it’s the global majority and that’s where the population growth is coming from, we consider that as a real key part of the sustainability of the open source ecosystem. And frankly, we hear a lot from developers in the Global North, and we just haven’t seen a lot in the way of data and actual insights on the perspectives and challenges of developers in the Global South. 

The sustainability of open source really hinges on having a strong pipeline of contributors and maintainers, so we’ve been asking ourselves at the Eclipse Foundation a number of really big questions, namely, where are the next generation of developers? Who are they? What’s their relationship with open source, and what challenges are they facing? 

And we started by looking at our own contributors and committers and research, including GitHub’s Octoverse report, and we noticed something that was interesting. The fastest growing developer communities are almost exclusively in the Global South, but what struck us more was that there wasn’t much research coming from these areas, and that’s why we decided to dive deeper. We wanted to understand the work they were doing, their perspectives, and also we had a hunch that the impact of open source was being felt far beyond just software development and also having broader socioeconomic effects.

Getting into the findings, 77% of respondents said they used open source software, 37% contribute to open source projects, 27% maintain them, and 22% create new projects. What has been the positive impact that these open source developers have been seeing?

The positive impact of these developers is something that we were positively surprised by. Three things in particular stood out for us in terms of that impact and the potential of these developers. 

First of all, they’re not just users of open source. They’re actively shaping its future. The fact that 28% are maintainers, and a quarter of them are creating new projects, it really means that they’re increasingly driving the agenda for these technologies that the rest of the world relies on. I think by now, pretty much everyone in the tech industry accepts that diversity is a good thing. I hope these developers are bringing their fresh perspectives and approaches and contributions to the communities that they’re part of. 

The second really big idea that we uncovered in terms of the positive impacts is that these developers are leveraging open source for career growth, very much like the rest of the world, and they’re using open source to acquire new skills, to learn new technologies and techniques and approaches to problem solving, and they’re also seeing that translate into better paying jobs and really seeing the financial benefits related to that. 

The third thing is that they’re also leveraging their involvement in open source to drive positive change more broadly in their communities. They identified three areas where they see that impact happening most greatly, and that’s improved educational opportunities for young people, for women, for other underrepresented groups in tech; the development of a stronger workforce overall in terms of software developers and those technology related functional areas; and  increased entrepreneurship, so business creation and economic contributions related to that innovation that’s based in software.

What are the ways in which they’re leveraging open source to advance their careers? And does it differ from how, for instance, developers in the US use open source to do that?

What we observed in our research is that there’s this democratizing effect that you see with open source because it’s permissionless. There’s no gatekeepers between someone in Johannesburg, South Africa, and a project that they want to use. It’s really removing and lowering barriers to access, and that’s huge.

Another thing is the fact that they’re able to leverage open source to build their skills, to advance their learning in a way that doesn’t require them to go to college or to university. That’s also really powerful, and it also extends to women. 

That’s something that we heard consistently, is that there are many countries in which women don’t have the same access to educational opportunities as they do in the in the West, and open source offers a very convenient and easily accessible way for those folks to get the skills they need to improve their lives, and then it also helps them collaborate with people from around the world. So you see this effect where technology, and particularly open source technology, really enables the borders of the world to come down, and people are able to relate to each other as community members.

You touched on the gender inequality part of this, the fact that women are able to kind of advance their careers better using open source, and the report pointed out that it’s also positive because open source solutions are being created that can impact gender specific issues, like apps for women’s specific healthcare issues or apps that provide like more educational resources. Can you share a little bit more about how open source is having that positive impact there, and also how policymakers can continue supporting women in open source in those countries?

I think one of the things that we did early on as we were developing the survey questionnaire is we talked to a number of experts, not only technology experts, but policymakers and folks who work at the UN, to get a better understanding of the Sustainable Development Goals, the SDGs, that factor into a lot of these larger questions around policy. They gave us some very helpful and concrete examples that we were able to test out in our quantitative research. 

Specifically, they said two things. They said that women are able to find mentors and role models and allies in these global open source communities. Again, this idea of being able to break through and go beyond the borders of their countries and those regions, and that’s the kind of thing that helps them build confidence. It reduces the sense of isolation and creates those new career opportunities through networking, especially in areas and countries where women are underrepresented in tech.

The second big idea was that women can contribute to open source projects that address the issues they care about. Now you mentioned healthcare and other applications that could be particularly targeted towards females and women. The idea that they can use their creativity, ingenuity, and passion to build solutions that work for everyone, not just a limited or small group of people, is really quite powerful, and that level of advocacy, and let’s say, focused enablement and participation, is the kind of thing that that really helps drive gender equality. 

We also heard a lot about how open source was empowering women and girls by offering them opportunities to better learn and contribute and lead in the tech industry. 

I’ll just share a bit of a small anecdote. At the Eclipse Foundation, we’ve actually partnered with the Girls Coding Academy in Lesotho — that’s the country that I’m from in southern Africa — and we’re working on an initiative with them to teach coding skills to about 200 teachers and girls in in that country using the Eclipse IDE. Now that’s just a small example. And we’re looking to do more along those lines, but hopefully that illustrates the fact that there’s a real connection between what is software and then how that software impacts people in their real day-to-day lives, and gives them opportunity and breaks down barriers that might otherwise exist.

Moving beyond the positive social impact. Another element of this report was that a majority think that open source is going to influence their country’s economic growth. Do you have any insights into why that is?

I think open source can help these countries drive economic growth in a few really important ways. One that we’ve already touched on is skill development and training. I think that’s a real way to help equalize and bridge the digital divide that exists between the Global South and North. The fact is that with open source, someone sitting in Palo Alto, California now has the same access to technology as someone sitting in Johannesburg, South Africa, or Lagos, Nigeria. And that’s a really an important shift in the world, frankly, and it allows for folks to unlock the often underused potential of a huge swath of the world. As we were talking about, this is the global majority, so now getting those people into the spheres of technology development and innovation is something that’s going to be beneficial, not only to these countries, but to the rest of the world. 

The other big impact is the fact open source enables startups and businesses to leverage technology to create opportunity. What we found is that developers in the Global South are having significant impact across a variety of industries, in existing businesses, in financial services, telecom, and healthcare, and increasingly, they’re creating new ventures. 

Over the last several years, we’ve seen startups be funded in Latin America and in Africa and in Asia from the Global North. We’re seeing these new ventures attract startup capital and interest and really advance the global technology scene from these countries.  So it’s not just a matter of outsourcing anymore and using these talented folks as a source of cheap labor, you’re actually seeing the developers and engineers from these regions make a mark on the international economy.

I know we’ve covered a lot of the highlights of the report, but were there any other takeaways from the report that developers might find interesting that we didn’t touch on?

I think maybe what I’d like to underline is that often, when we think of the Global South, you know, we think, how do we help these people? How do we assist them? And maybe the biggest takeaway for me was that this research shifts the narrative, where it’s not about what the rest of the world can do for the Global South, but particularly around the sustainability of open source, where you’re seeing a lot of the folks that created and maintained for many years these core technologies and infrastructure, you’re seeing those folks age out. So I think the narrative and discussion has shifted to now, how can the Global South help open source and help the tech industry? 

I think the fact is that with the leveling of the playing field that open source provides, and the fact that you’re seeing a lot of creative solutions and technologies come out of these countries, I would encourage developers in the Global North to look to their peers in the South as potential contributors, maintainers and leaders, and they should welcome and encourage and mentor them and ensure that they feel welcomed. That kind of engagement can help distribute the workload and reduce the burnout among maintainers today, and also inject new innovations into the global ecosystem.

The post The state of open source in the Global South appeared first on SD Times.

The OSI started the process of creating this definition back in 2022, and for the past year it has been traveling the world to gather feedback and further input on the draft definition. 

The group wrote in its announcement that this release was the result of “lots of community feedback: 5 town hall meetings, several comments on the forum and on the draft, and in person conversations at events in Austria, China, India, Ghana, and Argentina.” Now that this first release candidate is available, any future updates will only be bug fixes, not actual new features.

Generally, the Open Source AI Definition specifies that an AI system is open source if it meets the following criteria:

• Anyone can use it for any purpose without needing to ask permission
• Anyone can study how the system works or inspect its components
• Anyone can modify the system for any purpose
• Anyone can share the system with or without modifications, for any purpose.

According to the OSI, there have been three changes since the last release, all relating to the “preferred form to make modifications to a machine learning system.”

First and most notable is that there is new language around Data Information to clarify that training data must be shared and disclosed.

Second, it now specifies that code must be complete enough that downstream recipients can understand how training was done. “Training is where innovation is happening at the moment and that’s why you don’t see corporations releasing their training and data processing code. We believe, given the current status of knowledge and practice, that this is required to meaningfully fork (study and modify) AI systems,” the OSI wrote. 

Third, new text specifies that “it is admissible to require copyleft-like terms for any of the Code, Data Information and Parameters, individually or as bundled combinations.” For example, a consortium that owns the rights to training code and the dataset could distribute them both in a way that bundles them together with copyleft-like provisions. 

In its announcement, the OSI further reinforced the idea that the goal of open source (and also open source AI) isn’t to enable reproducible software, but rather to give anyone the ability to fork a system. 

“This is why OSD #2 requires that the “source code” must be provided in the preferred form for making modifications,” the organization wrote. “This way everyone has the same rights and ability to improve the system as the original developers, starting a virtuous cycle of innovation. Forking in the machine learning context has the same meaning as with software: having the ability and the rights to build a system that behaves differently than its original status. Things that a fork may achieve are: fixing security issues, improving behavior, removing bias. All these are possible thanks to the requirements of the Open Source AI Definition.”

Going forward, the OSI will be focusing on creating the Open Source AI Definition’s documentation, Checklist, and FAQ. The official 1.0 release is expected on October 28. 

The post Open Source AI Definition nears final version as first release candidate is announced appeared first on SD Times.

“Setting up a dev environment can feel like starting a car in the 1900s—engaging the handbrake, adjusting the fuel valve, mixture control, spark advance, choke, and throttle, turning the ignition, and often running into issues. With Daytona, it’s like starting a car in 2024: any driver can just push a button and go. Enabling developers to focus on what truly matters—writing code and building innovative solutions,” Ivan Burazin, CEO and co-founder of Daytona, wrote in a blog post. 

Development environments in Daytona are called Workspaces and they are reproducible, meaning that configurations and settings can be done once and then carried over. Currently, Workspaces are based on the Dev Container standard, but the project’s documentation claims that there is the potential to base it on other standards down the line, like Dockerfiles, Docker Compose, Nix, and Devfile. 

Daytona can run on any type of machine, including local, remote, cloud-based, physical service, VM, or any x86 or ARM architecture. 

It supports VS Code and JetBrains locally, and also has a built-in Web IDE. It also offers integrations with several Git providers, including GitHub, GitLab, Bitbucket, Bitbucket Server, Gitea, Fitness, Azure DevOps, and AWS CodeCommit.

Multiple project repositories can exist under one Workspace, so that developers using a microservices architecture can easily use Daytona for their development needs.

It also offers reverse proxy capabilities to enable collaboration among developers and streamline feedback loops.

For security purposes, during setup, it automatically creates a VPN connection from the client machine to the development environment. This connection also provides access to all ports in the development environment, which eliminates the need for setting up port forwarding.

According to a blog post written by Burazin, the project reached 4,000 stars on GitHub within the first week of the project being open sourced. Now it is at nearly 8,000 stars and has 39 developers contributing to it. 

The open-source project is built and maintained by a company of the same name, which in June received $5 million in seed funding to grow the project.

Read about other recent Open-Source Projects of the Week:

Teable | Penpot | Dioptra | Semantic Kernel’s Agent Framework | Hoppscotch

The post Daytona – SD Times Open Source Project of the Week appeared first on SD Times.

According to the organization back in May, there are many companies out there claiming their models are open source when they might really not be. Having a definition will make it easier for developers to make those determinations themselves. 

Draft version 0.0.9 further clarifies the components for Open Source models and Open Source weights, and states that all components of a system need to meet the open source standard in order for that system to be considered open source.

The OSI has also decided that training data won’t play a role in classification. “After long deliberation and co-design sessions we have concluded that defining training data as a benefit, not a requirement, is the best way to go,” the OSI wrote in a post. “Training data is valuable to study AI systems: to understand the biases that have been learned, which can impact system behavior. But training data is not part of the preferred form for making modifications to an existing AI system. The insights and correlations in that data have already been learned.” 

Some other changes in draft 0.0.9 are that the Checklist is now its own document, there are now references to conditions of availability of components, and the word “Model” was updated to “Weights” under the “Preferred form to make modifications,” because the way the word was used there was inconsistent with how it is used in the rest of the document. 

According to the OSI, the items still on the roadmap before October include continuing to improve the drafts based on feedback from meeting with shareholders at events around the world, updating the FAQ, establishing a review process for future versions of the definition, and deciding how to address reviews of new licenses for datasets, documentation, and agreements around model parameters. 

“Creating an Open Source AI Definition is an arduous task over the past two years, but we know the importance of creating this standard so the freedoms to use, study, share and modify AI systems can be guaranteed. Those are the core tenets of Open Source, and it warrants the dedicated work it has required,” OSI concluded.

The post OSI releases latest draft of Open Source AI Definition appeared first on SD Times.

Previously, CockroachDB was available as a free Core version that was designed more for individuals and startups, and a paid Enterprise version that was designed for larger companies and offered advanced capabilities for scaling. 

Starting November 18, 2024, CockroachDB Core will be retired and the company will only offer CockroachDB Enterprise, which will be free to individual developers, students, and companies that make less than $10 million in annual revenue. For companies that don’t qualify for free usage, there is now a 30 day free trial that they can use. 

These changes only apply to the self-hosted version; the cloud version will remain the same, the company claimed. 

According to Cockroach Labs, this change gives everyone — regardless of if they are paying customers — access to advanced features that were previously not available in Core, such as cluster optimization, disaster recovery, backup, streaming, and advanced security measures. 

The company explained that part of the reasoning behind this change is that two trends had emerged over the years. One is that startups and small businesses expressed an interest in using the advanced features not available in Core, and the second is that large companies would often compromise on using the full capabilities offered in Enterprise in order to stay on the free Core version. 

“In consolidating to a single CockroachDB Enterprise offering, we are ensuring that all of our users have access to the full breadth and scope of our product, delivered through a streamlined trial experience. Individuals and small businesses will now benefit from capabilities previously available only to paying customers, but continue using CockroachDB for free,” Spencer Kimball, CEO of Cockroach Labs, wrote in a blog post. 

The paid version of Enterprise will get access to support from Cockroach Labs, while the Free edition will only have access to support from the community, such as the Slack channel. 

Cockroach Labs explained that under this new license, CockroachDB will technically be “source code available.” 

These changes coincide with the release of CockroachDB 24.3, but will also apply to patches to CockroachDB 23.1 or later issued after that date.

The company has an FAQ on its website, and Kimball said it will continue adding to it as new information is available.

The post CockroachDB retires self-hosted Core offering, makes Enterprise version free for companies under $10M in annual revenue appeared first on SD Times.

Dioptra offers a common platform for assessing models throughout their life cycle, from when they are being developed to when they are acquired by other parties who want to then ensure trustworthiness. 

“Our systems increasingly rely on Machine Learning (ML) algorithms and models to perform essential functions,” NIST wrote in a post. “As users of these systems, we must implicitly trust that the models are working as designed. Establishing the trustworthiness of an ML model is especially hard, because the inner workings are essentially opaque to an outside observer.”

NIST defines several characteristics that a trustworthy AI model must have: “valid and reliable, safe, secure and resilient, accountable and transparent, explainable and interpretable, privacy-enhanced, and fair – with harmful bias managed.”

It offers several key features, including reproducibility of tests, traceability of tests, support for expanding functionality through plugins, interoperability among plugins, user authentication, an intuitive web interface, and the ability to be deployed in a multi-tenant environment where users can share and reuse components. 

It was designed to support NIST’s AI Risk Management Framework, which was announced in January 2023 to manage risks of AI to individuals, organizations, and society as a whole. Specifically, it fulfills the “Measure” component of the framework by providing tooling for assessing, analyzing, and tracking AI risk. 

“Initiatives like Dioptra are vital in ensuring AI technologies are developed and used ethically, reinforcing the commitment to safeguarding AI systems while promoting innovation,” said Michael Rinehart, vice president of artificial intelligence at Securiti, a company that also provides AI safeguards. 

Read about other recent Open-Source Projects of the Week:

• Semantic Kernel’s Agent Framework
• Hoppscotch
• DevPod

The post Dioptra – SD Times Open Source Project of the Week appeared first on SD Times.

The report recommends open models continue to be monitored for potential risks. Part of NTIA’s recommendation is that the government should start an ongoing program for collecting evidence of risks and benefits, evaluating the evidence, and then acting on it. 

The report came about because of President Biden’s 2023 Executive Order on “Safe, Secure, and Trustworthy Development and Use of Artificial Intelligence.” One of the things the EO did was instruct NTIA to conduct a report on the “potential benefits, risks, and implications of dual-use foundation models for which the model weights are widely available, as well as policy and regulatory recommendations pertaining to those models.”

Stefano Maffulli, executive director of the Open Source Initiative (OSI) and one of the primary stakeholders of the process, believes these recommendations will be an appropriate way for balancing risks and innovation. 

“There are undeniable benefits to research in open models, particularly in transparency around privacy, security and bias,” he said. “We encouraged equitable access to economic benefits by keeping models open, as this is a well-established factor in innovation globally.”

Dr. Thomas Di Giacomo, chief technology and product officer at SUSE, agrees that developing models in the open can help increase trust in AI. 

“SUSE believes that all software, including AI, should be freely accessed, used, changed and shared transparently by anyone … We believe community-based standards regarding AI will address the safety concerns of AI, as engaging the developer community is one of the best ways to improve security protocols and resilience,” he said. 

You may also like…

Open Source Initiative is close to coming up with a definition for Open Source AI

Open source in 2024: Tackling challenges related to security, AI, and long-term sustainability

The post White House is recommending that use of open source AI models not be restricted appeared first on SD Times.

Joining the board effective June 23, 2024 are John Gilmore, Christina Haralanova, and Maria Chiara Pievatolo. The board reviewed 83 nominations before selecting these three members. 

“The FSF has made a major step towards establishing a solid governance structure that sets us up for a strong future in which we can respond to software freedom threats ahead of us,” said Geoffrey Knauth, president of the FSF. 

The new members are the first people to join the board since 2020, according to the FSF. The organization also stated plans to have associate members review the current board members in August to decide if they should stay on board. 

“Given the importance of the FSF to the free software movement, and the importance of its board to ensure preservation of the software freedom definition, the board has not taken its task lightly,” the FSF wrote in a statement. 

You may also like…

Open source in 2024: Tackling challenges related to security, AI, and long-term sustainability

Forrester shares its top 10 emerging technology trends for 2024

The post Free Software Foundation adds three new board members appeared first on SD Times.

Companies can use it to govern structured and unstructured data, and machine learning models, notebooks, dashboards, and files. 

It offers interoperability with any data format and compute engine and supports all of the major cloud platforms. With Unity Catalog, companies can manage data from a number of sources in one place, including sources like MySQL, PostgreSQL, Amazon Redshift, Snowflake, Azure SQL, Azure Synapse, and Google BigQuery, to name a few.

Companies can define access policies once and then apply them across different clouds and platforms, simplifying access management and governance.  

Unity Catalog was first created at Databricks in 2021 as an offering for its customers, and it is being open sourced so that more companies can benefit from it. 

“We’re excited to open source Unity Catalog and release the code,” said Ali Ghodsi, co-founder and CEO of Databricks. “We’ll continue to evolve the open standard in close collaboration with our partners.”

Matt Dugan, VP Data Platforms, AT&T, added: “With the announcement of Unity Catalog’s open sourcing, we are encouraged by Databricks’ step to make lakehouse governance and metadata management possible through open standards. The flexibility to utilize interoperable tools with our data and AI assets, with consistent governance, is core to the AT&T data platform strategy.”

Read about other Open-Source Projects of the Week…

• OpenAI .NET library
• Apache Pinot, an analytics engine for large amounts of data
• Flow-IPC, a tool for making in-memory transfers of data, images, and videos between applications

The post Unity Catalog – SD Times Open Source Project of the Week appeared first on SD Times.