# Apache Projects for Small Business: Find What You Can Actually Use

Find useful Apache projects by business problem, see who must maintain them, and compare three years of hosting and labor with a paid service.

Author: J.A. Watte
Published: October 30, 2026
Source: https://jwatte.com/blog/apache-open-source-projects-small-business/

---

**Our catalog covers 327 current Apache directory entries: 303 established entries and 24 incubating projects. Seven are plausible projects for a capable small business owner or helper to run. Most are software you meet inside another product, or infrastructure your IT provider maintains.**

October 30, 2026 update: added setup time and costs when an AI agent does the typing.

Start with your problem. Need staff to stop answering the same question? Look at Answer or JSPWiki. Need ordinary desktop documents? Consider OpenOffice, after trying your actual files. Need to combine exported spreadsheets? Hop deserves a look. Buying servers because you recognize Kafka or Hadoop is a much harder business case.

The seven owner candidates are **Hop, Answer, HTTP Server, OpenOffice, JSPWiki, HertzBeat and StreamPipes**. That label assumes someone can follow technical instructions, handle updates and recover lost data. OpenOffice is a desktop application; the others involve more operational work. A container that starts successfully has not yet proved that your backups work.

Use the business table below, browse the [16 groups](#the-16-groups-and-the-count), or [download the complete catalog](/downloads/apache-projects-small-business-catalog.md) and search for a project name. Each group article explains its projects, the staffing involved and worked costs. This hub gives you the route into them.

We read official project material on October 29, 2026. We have not installed these projects for this series. Counts describe a directory snapshot, and the operating tiers are our judgments about a small business deployment. They are not ASF certifications or promises of current maintenance.

## Find your business first

These are possible starting points, not software shopping lists. The examples are illustrations. An existing business application may already do the job, and buying a second system can create another place to enter the same information. Every project below is either in our You tier or explicitly requires IT.

| Business and problem | Projects to consider | Why these fit |
|---|---|---|
| Nine person HVAC contractor whose office keeps answering installation questions | [Answer](/blog/apache-search-knowledge-base-small-business/), [JSPWiki](/blog/apache-websites-cms-web-apps-small-business/) | Put recurring questions in Answer or maintain approved procedures in a wiki, with someone responsible for corrections. |
| Six person landscaping crew joining scheduling exports to job costs | [Hop](/blog/apache-workflow-automation-small-business/), [OpenOffice](/blog/apache-email-documents-office-small-business/) | Try a repeatable import and a spreadsheet before commissioning a new management system. |
| Two location coffee shop comparing till totals | [Hop](/blog/apache-workflow-automation-small-business/), [Superset](/blog/apache-dashboards-analytics-small-business/) | Hop can prepare the exports; bring in IT for Superset when several managers need maintained dashboards. |
| Independent retailer with eight staff and an online store | [OFBiz](/blog/apache-erp-ecommerce-finance-small-business/), [Answer](/blog/apache-search-knowledge-base-small-business/) | Bring in IT for an integrated order and inventory system; Answer can organize public product questions separately. |
| Five person auto repair shop documenting recurring workshop procedures | [JSPWiki](/blog/apache-websites-cms-web-apps-small-business/), [OpenOffice](/blog/apache-email-documents-office-small-business/) | Keep shared procedures searchable and prepare ordinary forms without turning the wiki into a repair billing system. |
| Property manager with four office staff and repeated onboarding questions | [Answer](/blog/apache-search-knowledge-base-small-business/), [Hop](/blog/apache-workflow-automation-small-business/) | Answer holds general procedures while Hop prepares routine exports, with tenant records kept in the designated business system. |
| Self storage operator with three sites checking website availability | [HertzBeat](/blog/apache-monitoring-logging-small-business/), [Guacamole](/blog/apache-security-identity-remote-access-small-business/) | Monitor the sites and bring in IT if browser based remote access to office computers is justified. |
| Twenty room motel with six staff sharing shift instructions | [JSPWiki](/blog/apache-websites-cms-web-apps-small-business/), [OpenOffice](/blog/apache-email-documents-office-small-business/) | Store operating instructions and printable checklists while keeping bookings and payments in the existing reservation system. |
| Two site laundromat collecting machine readings | [StreamPipes](/blog/apache-iot-edge-devices-small-business/), [HertzBeat](/blog/apache-monitoring-logging-small-business/) | Explore a small, isolated device data pilot and monitor its services; involve IT before connecting production machine networks. |
| Car wash with two sites and compatible equipment telemetry | [StreamPipes](/blog/apache-iot-edge-devices-small-business/), [IoTDB](/blog/apache-iot-edge-devices-small-business/) | Try visual sensor flows; bring in IT if keeping and querying device history warrants a dedicated database. |
| Twelve person machine shop reconciling orders and production | [OFBiz](/blog/apache-erp-ecommerce-finance-small-business/), [Hop](/blog/apache-workflow-automation-small-business/) | Bring in IT for OFBiz configuration and migration, and assess Hop for bounded export jobs. |
| Seven person print shop repeating file preparation instructions | [Answer](/blog/apache-search-knowledge-base-small-business/), [HTTP Server](/blog/apache-web-servers-api-gateways-small-business/) | Publish reusable questions and answers, or serve a simple informational website with a capable administrator. |
| Nonprofit with four staff and rotating volunteers | [JSPWiki](/blog/apache-websites-cms-web-apps-small-business/), [OpenOffice](/blog/apache-email-documents-office-small-business/) | Maintain volunteer instructions and ordinary documents; first check compatibility with funder templates. |
| Farm or winery with six staff comparing production records | [Hop](/blog/apache-workflow-automation-small-business/), [StreamPipes](/blog/apache-iot-edge-devices-small-business/) | Start with repeatable file preparation, then consider a small sensor pilot only where compatible equipment provides usable readings. |

The group articles carry the costs and limitations. No row promises an industry specific package. A wiki does not become a reservation system, and a dashboard does not reconcile the accounts that feed it.

## What the Apache Software Foundation actually does

The [Apache Software Foundation](https://www.apache.org/foundation/) is a nonprofit home for open source communities. Its governance page describes a member based organization whose members elect a board. Project Management Committees, usually shortened to PMCs, oversee individual projects. Contributors and committers do the work through project communities, with public discussion and project decisions rather than a single commercial product roadmap. [ASF governance](https://www.apache.org/foundation/governance/) explains those responsibilities.

For an owner, that distinction affects where help comes from. Downloading Apache software does not buy an implementation team, a response time or someone who will restore your database tonight. You can hire those services separately when a qualified provider offers them. Your provider's contract, the project's release process and your own recovery arrangements answer different questions.

The [Incubator](https://incubator.apache.org/) helps incoming projects develop an Apache community and meet the foundation's requirements. We label every such entry **(Incubating)**. The [Incubator disclaimer guidance](https://incubator.apache.org/guides/branding.html) explains that these projects are not yet fully endorsed by the ASF; incubation does not guarantee graduation. That is a reason to investigate continuity and support before committing business records, not an automatic rejection of the software.

The [Attic](https://attic.apache.org/) preserves information about retired Apache projects. Its oversight committee is "not authorized to actively develop and release" those projects. Historical downloads can remain available after maintenance ends. Also read the project's own status notice: a retired subproject can still have a familiar name and an old download page.

An Apache name identifies the project and its relationship to the foundation. It does not promise that a product fits your business, that every component is maintained, or that a commercial vendor is endorsed. The [ASF trademark policy](https://www.apache.org/foundation/marks/) treats names and logos separately from permission to use the code.

In plain terms, [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0) permits commercial use, modification and redistribution, subject to its terms. Redistribution has requirements around the license, notices and identification of changes; the license also includes a contributor patent grant with conditions. It does not grant general trademark rights or a warranty. This is a reading guide, not legal advice. Ask counsel about your distribution, and check bundled dependencies rather than assuming every file has identical terms.

Security notices belong to the individual projects. Start at [ASF security](https://www.apache.org/security/), then follow the project's security page and advisory instructions. The ASF page separates reporting a suspected vulnerability from ordinary support. A supplier that embeds Apache components should tell you how it tracks those advisories and delivers fixes to your installed product.

## The 16 groups and the count

We classify entries by the work they do, then by who should operate or integrate them. **You** means a capable owner or helper can handle the limited use described. **Your IT department or provider** means ongoing operational skills are needed. **Your developer** means code, build tools or contributor resources rather than an application the owner should install.

The final column is an orientation list. IT and developer picks are things to discuss with those people, not additions to the seven owner candidates.

| Group and contents | Entries | You | IT department or provider | Developer | Worth recognizing first |
|---|---:|---:|---:|---:|---|
| [1. ERP, ecommerce and finance](/blog/apache-erp-ecommerce-finance-small-business/): orders, stock, customer data and financial back offices | 3 | 0 | 3 | 0 | OFBiz and Fineract, with IT |
| [2. Dashboards and analytics](/blog/apache-dashboards-analytics-small-business/): reporting interfaces and analytical databases | 11 | 0 | 7 | 4 | Superset and Doris, with IT |
| [3. Workflow automation](/blog/apache-workflow-automation-small-business/): scheduled jobs and moving data between systems | 11 | 1 | 7 | 3 | Hop; Airflow and NiFi with IT |
| [4. Search and knowledge](/blog/apache-search-knowledge-base-small-business/): questions, answers and document search | 10 | 1 | 4 | 5 | Answer; Solr with IT |
| [5. Messaging and events](/blog/apache-messaging-event-streaming-small-business/): reliable delivery between applications | 11 | 0 | 11 | 0 | ActiveMQ, Artemis and Kafka, with IT |
| [6. Web servers and API gateways](/blog/apache-web-servers-api-gateways-small-business/): serving pages and routing application requests | 8 | 1 | 6 | 1 | HTTP Server; Tomcat and APISIX with IT |
| [7. Databases](/blog/apache-databases-small-business/): storing application records and operational data | 19 | 0 | 16 | 3 | CouchDB and Cassandra, with IT |
| [8. Big data and lakehouse](/blog/apache-big-data-lakehouse-small-business/): processing large datasets and their file formats | 46 | 0 | 29 | 17 | Spark and Flink with IT; Parquet through a developer |
| [9. Security and remote access](/blog/apache-security-identity-remote-access-small-business/): accounts, permissions and remote desktops | 15 | 0 | 7 | 8 | Guacamole and Syncope, with IT |
| [10. Email, documents and office](/blog/apache-email-documents-office-small-business/): desktop documents, mail and file handling | 10 | 1 | 3 | 6 | OpenOffice; SpamAssassin and James with IT |
| [11. Websites and web applications](/blog/apache-websites-cms-web-apps-small-business/): wikis, publishing and application frameworks | 22 | 1 | 3 | 18 | JSPWiki; Roller with IT |
| [12. Private cloud and clusters](/blog/apache-private-cloud-cluster-tools-small-business/): allocating and coordinating computing resources | 14 | 0 | 9 | 5 | CloudStack and ZooKeeper, with IT |
| [13. Monitoring and logging](/blog/apache-monitoring-logging-small-business/): availability, application traces and recorded events | 11 | 1 | 3 | 7 | HertzBeat; SkyWalking with IT; Log4j through a developer |
| [14. Devices and sensors](/blog/apache-iot-edge-devices-small-business/): equipment readings and embedded software | 7 | 1 | 1 | 5 | StreamPipes; IoTDB with IT |
| [15. AI and machine learning](/blog/apache-ai-machine-learning-small-business/): model tooling and language processing | 12 | 0 | 1 | 11 | OpenNLP and TVM through a developer |
| [16. Developer libraries and build tools](/blog/apache-developer-libraries-build-tools-small-business/): reusable code, testing, builds and contributor resources | 117 | 0 | 1 | 116 | Maven, NetBeans and Subversion through a developer |
| **Total** | **327** | **7** | **111** | **209** | |

### Why this is 327, rather than 310

The [project registry](https://projects.apache.org/json/foundation/projects.json) snapshot read October 29, 2026 at 21:00 UTC contains 380 records. Removing its 70 records categorized as retired leaves 310. That is a valid calculation of that file, but it misses part of what the [directory page](https://projects.apache.org/projects.html) displays.

The directory merges the [podling list](https://projects.apache.org/json/foundation/podlings.json) with the project registry, with registry records taking precedence for duplicate IDs. There are 24 podlings and one overlapping PouchDB record, giving 403 displayed entries before exclusions. We remove the 70 categorized retirements and six additional entries with explicit official retirement notices: IvyDE, the .NET and VSS Ant libraries, Annotator, Woden and Derby. The result is **403 minus 70 minus 6 = 327**, or **310 plus 23 minus 6 = 327**.

The directory union and retirement checks were completed at 21:15:14 UTC on October 29. The download gives the sources for those exclusions. Its 303 established entries include components within larger projects and contributor resources, alongside 24 podlings. This is not a count of 327 separate ready made business applications.

We retained directory entries marked dormant, inactive or end of life when they did not meet the explicit retirement exclusion. Their catalog warnings matter. Directory membership alone does not establish active maintenance. Release information is the latest found in the cited sources, which sometimes means registry metadata, a representative component or a distribution file date. The catalog distinguishes those cases instead of treating every date as a fresh release announcement.

## Check these things before relying on a project

1. **Read the latest release page.** Check the version you would actually install, its date and supported release line. An old date deserves investigation; it does not explain the reason by itself.
2. **Look at the release cadence.** Compare several releases and their notes. Frequent releases can mean active development, but you still need a manageable update process.
3. **Find its security page.** Identify where advisories appear, who receives them at your business and how quickly you can patch. Check dependencies too.
4. **Read recent mailing list discussion.** Look for answered operating questions, release discussions and evidence that more than one person can maintain the project. Silence is a question to resolve before purchase.
5. **Find actual paid support.** Use the project's vendor references or the provider's own description. Ask what versions, hours and restoration work the contract covers. A logo on a website is not a service agreement.
6. **Read the production architecture.** Does the useful deployment need one machine, a separate database or a cluster? Ask what survives a failed node, how backups are restored and who tests that recovery.

For any shortlisted system, write down the person who owns these checks. If everyone assumes the installer will do them later, you have an unassigned job before the first user signs in.

## How every cost example works

The software license price is only one line. We compare **36 months** of setup, hosting, backups and administration with a named paid alternative. Prices below are US dollars before tax, read October 29, 2026. Server figures are monthly caps without introductory credits; the comparison service states its billing term separately.

| Shared input | Value used | Basis and limits |
|---|---|---|
| Server administration labor | $68.09 an hour | [BLS May 2025 administrator median wage](https://data.bls.gov/oesprofile/?areas=INDUSTRY,STATE,MSA&major_group=150000&measure=01&occupation=151244&year=2025), $47.66, divided by the [June 2026 wages share](https://data.bls.gov/timeseries/CMU2020000000000P), 0.700 |
| User support labor, only for appropriate work | $42.49 an hour | [BLS May 2025 support specialist median wage](https://data.bls.gov/oesprofile/?areas=INDUSTRY,STATE,MSA&major_group=150000&measure=01&occupation=151232&year=2025), $29.74 / 0.700; not the server rate |
| Small server | $24 a month | [DigitalOcean Basic Regular](https://www.digitalocean.com/pricing/droplets), 2 shared vCPUs, 4 GiB RAM, 80 GiB disk, NYC3 |
| Medium server | $48 a month | Same provider, Basic Regular, 4 shared vCPUs, 8 GiB RAM, 160 GiB disk |
| Large server | $252 a month | Same provider, General Purpose Regular, 8 dedicated vCPUs, 32 GiB RAM, 100 GiB disk |
| Off site backup storage | $6.95 per decimal TB per month | [Backblaze B2](https://www.backblaze.com/cloud-storage/pricing), first 10 GB free; count all retained copies |
| One server setup and monthly care | 4 to 12 hours setup; 1 to 3 hours a month | Our planning assumption for a limited owner project |
| Small cluster setup and monthly care | 24 to 60 hours setup; 6 to 16 hours a month | Our planning assumption with IT or provider oversight; multiply hosting by actual nodes |

The labor calculation estimates employee compensation with benefits. It is not an IT company's billing rate. An outside provider will normally quote more because its fee also covers overhead and other costs. Neither unpaid owner time nor a helpful relative makes maintenance disappear.

Backup storage costs `max(0, stored GB minus 10) x $0.00695` per month. B2 includes outbound transfer up to three times average stored data; extra transfer is $0.01 per GB. Classes A, B and C calls are free for pay as you go customers. Chargeable [Class D calls](https://www.backblaze.com/cloud-storage/transaction-pricing) cost $0.004 per 10,000 after the first 2,500 each day. Add source host transfer charges where applicable. Storage payment buys space; checking backups and restoring them consumes labor.

Use these formulas with unrounded inputs, then round displayed totals to whole dollars:

- Setup = setup hours x loaded hourly rate.
- Monthly = all server charges + backup storage and usage + monthly admin hours x loaded hourly rate.
- Self hosted total = setup + 36 x monthly.
- Paid alternative total = its setup cost + 36 x verified monthly price.
- Savings = paid alternative total minus self hosted total.
- Break even month = round up ((self hosted setup minus alternative setup) / (alternative monthly cost minus self hosted monthly cost)), when the monthly saving is positive; minimum zero.

If self hosting has equal or higher monthly costs and no setup advantage, break even is **never**. For a cluster, the production documentation determines the node count. A small server price is not a license to pretend a distributed system needs only one node.

Our AI assisted planning assumptions retain 50% to 80% of manual setup hours and 75% to 100% of monthly admin hours for one server; clusters retain 65% to 90% and 80% to 100%, respectively. Pair low with low and high with high. These are assumptions, not measured Apache installations or study results. Prompting, supervision, review, testing and corrections are included in those hours; unattended agent time is not automatically human labor. Hosting and backups stay unchanged, and tool costs are added separately. Coding studies report both gains and slowdowns, so they do not establish a deployment saving. See sources 10 to 16 and the worked example below.

### Worked example: a small public support community

For example, a ten person print shop wants customers to find answers about file preparation and mark a helpful reply as the accepted answer. It considers [Apache Answer](https://answer.apache.org/docs/installation/) on one server and hosted [Discourse Pro](https://www.discourse.org/pricing). Discourse is a broader forum; this comparison is for the overlapping community support job, not identical software.

Answer's official installation guide recommends Docker Compose and describes SQLite as the smallest database environment. It gives no CPU or RAM minimum. The shared 4 GiB server is our planning allowance, and we would test the workload before committing. The operator still owns access, persistent data, updates and recovery. Discourse Pro is listed at **$100 a month**, billed monthly, including the Solved feature, custom domains, five admin or moderator seats and unlimited members. We assume two moderators. Its 20 GB upload allowance is not the same measure as retained backup volume.

Our assumptions are one server and ten staff in both low and high cases, with two moderators in both; customer traffic remains a pilot workload to measure. Retained backup storage is 50 GB low and 100 GB high, including versions, with the free 10 GB allocation otherwise unused. Setup is 4 hours low or 12 high; administration is 1 hour a month low or 3 high. We assume no chargeable backup transfer or Class D calls. Paid service setup is zero additional technical hours in this simplified comparison; content writing and moderation are excluded from both sides, as are migration and training. An existing domain and any existing notification email allowance add no charge in this illustration; price extra services before buying.

| Cost over 36 months | Answer, low case | Answer, high case | Discourse Pro |
|---|---:|---:|---:|
| Setup | $272 | $817 | $0 assumed |
| Monthly server or subscription | $24 | $24 | $100 |
| Monthly backup storage | $0 after rounding | $1 after rounding | No separate backup cost assumed; verify retention and exports |
| Monthly administration labor | $68 | $204 | No separate technical labor assumed |
| Monthly total | $92 | $229 | $100 |
| **36 month total** | **$3,598** | **$9,057** | **$3,600** |
| **Savings from self hosting** | **$2** | **Negative $5,457** | Comparison baseline |
| Break even | Month 36 | Never | |

The low calculation is `4 x 68.09 + 36 x (24 + 40 x 0.00695 + 1 x 68.09) = 3597.608`. The high calculation is `12 x 68.09 + 36 x (24 + 90 x 0.00695 + 3 x 68.09) = 9057.318`. Discourse is `36 x 100 = 3600`. Low monthly cost is $92.368 before rounding, so break even is the ceiling of `272.36 / (100 minus 92.368)`, or month 36. The backup line rounded to zero is still charged in the calculation.

**Self hosting has no meaningful financial win in the low case and loses clearly in the high case.** A single unbudgeted support task can erase that two dollar difference. Discourse also lists a free plan; assess that first if its limits and features meet your needs. Choosing Answer could still make sense for its particular question and answer experience or control over data, but those are reasons to value, not savings to invent.

## Setup time when Claude or ChatGPT does the typing

We keep the print shop's Answer server at $24 monthly and retained backups at 50 GB low or 100 GB high. Our assumed hours include 1 to 3 hours of setup review and 0.5 to 1.5 hours of monthly review, already inside the AI totals. No installation was timed.

| Answer planning input or 36 month total | Low case | High case |
|---|---:|---:|
| Manual setup hours | 4 | 12 |
| AI assisted setup hours | 2 | 9.6 |
| Manual monthly admin hours | 1 | 3 |
| AI assisted monthly admin hours | 0.75 | 3 |
| Manual total | $3,598 | $9,057 |
| Claude Pro subscription total | $3,049 | $9,094 |
| ChatGPT Plus subscription total | $3,049 | $9,094 |
| API total: Claude Opus 5.5 | $2,863 | $9,008 |
| API total: GPT-6 Astra | $2,886 | $9,179 |
| Discourse Pro total | $3,600 | $3,600 |

Claude Pro and ChatGPT Plus each cost $20 monthly. We charge a separate setup month, then allocate 25%, or $5 monthly, for 36 operating months only when other work uses the remainder. With no other use, add $540. Equal prices do not mean equal allowances; check limits and budget overruns. Optional Desktop Commander Remote Pro adds $200 at that allocation; the base uses free local tooling. Sources 10, 11 and 13 document prices.

Our cumulative API assumptions are 200,000 to 1,000,000 setup input tokens and 20,000 to 100,000 output; monthly input is 100,000 to 500,000 and output 10,000 to 50,000. Low input splits 30% ordinary, 10% short cache writes, 60% reads; high has no caching. Retries and billed reasoning are included. Source 12 gives rates; our resulting setup/monthly token charges are $0.764/$0.382 low and $6/$3 high for Opus, $1.97/$0.985 and $15/$7.50 for Astra.

Use `manual = S x L + 36 x (H + A x L)` and `AI = S x q x L + T0 + 36 x (H + A x r x L + Tm)`. Here S/A are manual setup/monthly hours, q/r their retained shares, L is $68.09, H is hosting plus backups, and T0/Tm are setup/monthly tools. Round only final totals.

Claude Code or Codex can draft Answer's Compose configuration and restore commands. Desktop Commander on the server or through authorized SSH can install Docker, edit configuration and read Answer logs. Claude in Chrome can inspect Cloudflare DNS, the provider panel and Answer's admin UI; ChatGPT's verified cloud browser uses a separate browser session. Sources 10, 11, 13 and 14 describe these tools.

Use a new server, keep secrets out of chat, and approve commands or use a restricted sandbox. Watch dashboard actions, review changes and test an Answer restore; we still own backups and updates. Use subscriptions for people directing supported tools and API keys for custom automation, subject to each provider's terms in source 15.

**The low conclusion changes to modest savings:** subscriptions break even against Discourse in month 8, APIs in month 6. Charging the full subscription leaves only $11 saved over 36 months. Every high case still loses; its unchanged admin hours leave recurring tool charges. The high subscription case also exceeds manual Answer costs from month 29.

## Getting help without inheriting an unstaffed job

Do it yourself when the job is bounded, the documentation is understandable and you can demonstrate a restore before real users depend on it. Start with nonessential information. Put updates and backup checks on the calendar, and record how another person can take over. For OpenOffice, the first test is much simpler: open representative files, edit them and send them back to the people who need them.

Pay your IT provider when the project handles remote access, business databases, mail delivery or a cluster. Ask for a written scope covering installation, routine updates, monitoring, backup retention and recovery testing. Establish who receives security notices and what happens outside normal hours. A working demonstration and a maintained business service are different deliverables.

Stay on SaaS when the subscription costs less than the time you can responsibly assign, or when nobody wants ownership of the underlying system. Also compare support, exports, account recovery and contract terms. Paying a subscription does not remove your responsibility for permissions, staff departures and keeping usable copies of your records.

Keep the first project small enough to reverse. For the print shop, that means copying a limited set of public instructions into a trial community, testing search with staff and checking an export before customers depend on it. Do not migrate the order system just because the discussion board worked. Expanding only after a useful result gives you a better basis for the next purchase.

## Download the catalog

[Download the complete Apache small business catalog in Markdown](/downloads/apache-projects-small-business-catalog.md). It contains all 327 entries once each, grouped by purpose, with homepage links, operating tiers, release evidence and status cautions. Save it, search it or give it to your provider with two or three questions circled. The useful output is a short decision, not a plan to install the whole directory.

## Related reading

- [A curated Apache stack for small and medium businesses](/blog/apache-stack-small-medium-business/) connects the pieces, prices three years of operation and explains when Flink and Iceberg earn a place.
- [A small development server behind Cloudflare email login](/blog/kimsufi-vps-cloudflare-tunnel-access/) shows how to keep staging copies of any of these projects off the open internet, on a Kimsufi box or a small VPS.
- [Laravel for small business](/blog/laravel-small-business-guide/) covers custom web apps built on PHP, what they cost against SaaS, and hosts that support them, including a Laravel image that runs on Apache HTTP Server.
- [n8n for small business](/blog/n8n-small-business-automation-guide/) looks at another way to connect routine tasks. n8n is outside this Apache catalog.
- [Local AI versus cloud costs](/blog/local-ai-on-prem-vs-cloud/) discusses another purchase where administration and hardware change the apparent savings.
- [Choose your business web stack](/blog/blog-build-your-web-stack-visualizer/) helps place a website, data store and business tools in the same picture.

For a broader small business launch plan, [The $97 Launch](/books/) is related reading. The examples here stand on their own; no purchase is needed to use the catalog.

## Fact-check notes and sources

The catalog research was completed October 29, 2026. Release details reflect the sources read on that date and do not promise support. Exact price observation times below refer to the shared research inputs. Project descriptions and operating judgments are documented in the download and linked group articles.

1. [ASF project directory](https://projects.apache.org/projects.html), [project registry](https://projects.apache.org/json/foundation/projects.json) and [podlings](https://projects.apache.org/json/foundation/podlings.json): registry snapshot at 21:00 UTC; union and exclusion review at 21:15:14 UTC, October 29, 2026. The catalog lists the six additional retirement sources.
2. [Foundation](https://www.apache.org/foundation/) and [governance](https://www.apache.org/foundation/governance/): read October 29, 2026 at 21:34:23 and 21:34:24 UTC. Member, board and PMC responsibilities.
3. [Incubator](https://incubator.apache.org/), [disclaimer guidance](https://incubator.apache.org/guides/branding.html) and [Attic](https://attic.apache.org/): read October 29, 2026 at 21:34:26, 21:37:44 and 21:34:27 UTC respectively. Incubation and retirement status.
4. [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0), [trademarks](https://www.apache.org/foundation/marks/) and [security](https://www.apache.org/security/): read October 29, 2026 at 21:34:28, 21:34:29 and 21:34:31 UTC respectively.
5. [BLS administrator wage](https://data.bls.gov/oesprofile/?areas=INDUSTRY,STATE,MSA&major_group=150000&measure=01&occupation=151244&year=2025), [support wage](https://data.bls.gov/oesprofile/?areas=INDUSTRY,STATE,MSA&major_group=150000&measure=01&occupation=151232&year=2025) and [private industry wages share](https://data.bls.gov/timeseries/CMU2020000000000P): read October 29, 2026 at 21:12:08, 21:12:18 and 21:17:59 UTC. These supply the employee compensation estimates, not vendor quotes.
6. [DigitalOcean Droplet prices](https://www.digitalocean.com/pricing/droplets) and [regional availability](https://docs.digitalocean.com/platform/regional-availability/): read October 29, 2026 at 21:10:42 and 21:13:07 UTC. NYC3 is the shared comparison region.
7. [Backblaze B2 prices](https://www.backblaze.com/cloud-storage/pricing): read October 29, 2026 at 21:10:42 UTC. Storage, transfer and request allowances feed the backup formula.
8. [Answer installation](https://answer.apache.org/docs/installation/): read October 29, 2026 at 21:34:33 UTC. Installation recipe and database choices; no installation test was performed.
9. [Discourse pricing](https://www.discourse.org/pricing): read October 29, 2026 at 21:37:31 UTC. Pro monthly pricing and included features; plan limits should be checked again before purchase.
10. [Claude pricing](https://claude.com/pricing): read October 29, 2026 at 23:31:08 UTC. Pro is $20 billed monthly and includes Claude Code; annual prepayment is a separate cash commitment.
11. [ChatGPT pricing](https://chatgpt.com/pricing/): read October 29, 2026 in a regular browser. Plus is $20 monthly. [Codex plan access](https://help.openai.com/en/articles/11369540-using-codex-with-your-chatgpt-plan), read at 23:31:16 UTC, and [Codex CLI](https://developers.openai.com/codex/cli), read at 23:32:01 UTC that day, describe supported sign-in and server tooling.
12. [Anthropic API pricing](https://platform.claude.com/docs/en/about-claude/pricing), [OpenAI API pricing](https://developers.openai.com/api/docs/pricing) and [OpenAI caching](https://developers.openai.com/api/docs/guides/prompt-caching): read October 29, 2026 at 23:31:14, 23:31:13 and 23:32:00 UTC. Per million tokens, ordinary input/output/short cache write/cache read rates are $4/$20/$5/$0.20 for Opus 5.5 and $10/$50/$12.50/$1 for GPT-6 Astra. Token cost = (ordinary input x its rate + writes x their rate + reads x their rate + output x its rate) / 1,000,000. These are standard synchronous rates within the applicable context band; no batch discount or conditional credits are deducted.
13. [Desktop Commander pricing](https://desktopcommander.app/pricing/), [device access](https://desktopcommander.app/) and [local server documentation](https://raw.githubusercontent.com/wonderwhy-er/DesktopCommanderMCP/main/README.md): read October 29, 2026 at 23:31:08, 23:31:09 and 23:31:41 UTC. Local tooling is free; optional Remote Pro is $20 monthly. Model tokens and hosting are separate. Command restrictions alone are not a sandbox.
14. [Claude in Chrome](https://support.claude.com/en/articles/12012173-get-started-with-claude-in-chrome) and [permissions](https://support.claude.com/en/articles/12902446-claude-in-chrome-permissions-guide): read October 29, 2026 at 23:31:11 and 23:31:09 UTC. No separate extension charge is allocated. [ChatGPT cloud browser](https://help.openai.com/en/articles/20001280-using-cloud-browser-in-chatgpt), read at 23:32:01 UTC that day, describes availability subject to rollout and workspace controls; it does not install server packages.
15. [Anthropic Consumer Terms](https://www.anthropic.com/legal/consumer-terms), read October 29, 2026 at 23:31:08 UTC, restrict automated or nonhuman access except through an API key or where explicitly permitted. [OpenAI Terms](https://openai.com/policies/terms-of-use/), read that day in a regular browser, prohibit programmatic extraction of data or Output and credential sharing. These provisions do not prohibit every supported agent feature; check the documented feature and applicable agreement.
16. [Controlled Copilot experiment](https://arxiv.org/abs/2302.06590), published February 13, 2023, read October 29, 2026 at 23:31:08 UTC; [METR experienced developer study](https://metr.org/blog/2025-07-10-early-2025-ai-experienced-os-dev-study/), published July 10, 2025, read October 29, 2026 at 23:30:58 UTC. Different coding tasks produced different results; neither measures Apache server setup.

*Apache, the Apache feather logo and the Apache project names are trademarks of The Apache Software Foundation. This site is not affiliated with or endorsed by the ASF. Other product names are used for identification only.*


---

Canonical HTML: https://jwatte.com/blog/apache-open-source-projects-small-business/
RSS: https://jwatte.com/feed.xml
JSON Feed: https://jwatte.com/feed.json
Hero image: https://jwatte.com/images/apache-open-source-projects-small-business.webp